• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention is a method of processing a pulse (S) generated by a detector (10) of ionizing radiation, the detector (10) being able to interact with ionizing radiation (5) to generate said pulse, of which the amplitude (A) depends on an energy released by the ionizing radiation during its interaction in the detector, the method comprising the following steps: a) exposing the detector to an ionizing radiation source (2) so as to obtain, at a measuring instant (t), a pulse (S), called measuring pulse; b) shaping of the measuring pulse, considering a first shaping time (dt1), and determining a first amplitude (A of the pulse thus shaped, c) correcting the first amplitude measured in step b), taking into account a correction factor (?); the correction factor being determined by taking into account pulses, called pulses of interest (S1 .... SN, Scalib), formed by the detector during exposure to the source or a calibration source, during a time range of interest (At).
    公开号:FR3069066A1
    申请号:FR1756775
    申请日:2017-07-17
    公开日:2019-01-18
    发明作者:Andrea Brambilla；Cinzia De Cesare
    申请人:Commissariat a lEnergie Atomique CEA；Commissariat a lEnergie Atomique et aux Energies Alternatives CEA；
    IPC主号:
    专利说明:

    Method for processing a pulse generated by an ionizing radiation detector Description
    TECHNICAL AREA
    The invention relates to ionizing radiation detectors, in particular X or gamma photon radiation.
    PRIOR ART
    Ionizing radiation detection devices, based on gaseous, semiconductor or scintillator detector materials, make it possible to obtain electrical pulses formed by interactions of the radiation in the detector material. The amplitude of each pulse depends on the energy deposited by the radiation during each interaction. These devices are frequently used in applications requiring knowledge of the energy of the ionizing radiation incident to the detector. The fields of application are wide, and in particular include non-destructive testing, for example baggage screening, medical diagnosis or measurements in nuclear installations. Generally, the devices comprise an electronic circuit for processing the pulses allowing an estimation as precise as possible of their amplitudes. In particular, a circuit for shaping the pulses generated by the detector is used, so that the amplitude of each pulse can be estimated with precision. Indeed, on this precision depends the precision with which the energy of the radiation is estimated, which conditions the energy resolution of the detector. The shaping circuits are generally parameterized by a time constant, or shaping duration, according to which each pulse is analyzed. It is generally agreed that the optimal time constant is greater than or equal to the rise time of the electrical pulses formed by the detector.
    Certain applications require carrying out measurements known as high counting rates, when the detector is exposed to intense incident radiation. In such configurations, two interactions, originating from different radiations, can occur at the same instant, or in a time interval so short that the electrical pulse formed is no longer representative of an interaction formed by a radiation, but by a superposition of the pulses formed consecutively by different radiations. This corresponds to a stack. The amplitude of a pulse resulting from a stack is high, because it corresponds to the sum of all or part of the energies deposited in the detector by several radiations. Thus, a pulse resulting from a stack is useless for determining the energy of the incident radiation.
    In order to limit the occurrence of stacks, it is recognized that it is necessary to reduce the shaping time according to which the shaping circuit processes each pulse, to the detriment of the energy resolution, as mentioned in the publication Goulding FS Large coaxial germanium detectors - Correction for bailistic deficit and trapping losses, IEEE transactions on nuclear science, Vol.37, No2, April 1990. In the aforementioned publication, the authors use two different shaping circuits, treating each pulse respectively according to a first shaping time and according to a second shaping time, the second shaping time being longer than the first shaping time. The detector used is a large volume Germanium type coaxial detector, this type of detector being particularly suitable for high resolution measurements in low flux energy. This method includes an estimation of the ballistic deficit of each pulse processed by the second shaping time, that is to say the longest shaping time. The estimation of the ballistic deficit is estimated by combining a subtraction of the respective amplitudes of the pulses resulting from the two shaping circuits, with a ratio between the two durations of shaping. The ballistic deficit thus estimated is added to the pulse processed by the second shaping circuit, the main objective being to optimize the energy resolution.
    The inventors of the present invention have estimated that the method described above is difficult to apply in situations in which a small volume detector exposed to high counting rates is used. They propose a method for processing a pulse generated by a detector, compatible with use in high counting rates, and making it possible to maintain an estimate of the energy, corresponding to each interaction, reliable and robust.
    STATEMENT OF THE INVENTION
    A first object of the invention is a method for processing a pulse generated by an ionizing radiation detector, the detector being able to interact with ionizing radiation to generate said pulse, the amplitude of which depends on an energy released by ionizing radiation during its interaction in the detector, the process comprising the following steps:
    a) exposure of the detector to a source of ionizing radiation so as to obtain, at a measurement instant, a pulse, called measurement pulse;
    b) shaping of the measurement pulse, considering a first shaping time, and determination of a first amplitude of the measurement pulse thus shaped;
    c) correction of the first amplitude determined during step b), taking into account a correction factor;
    the correction factor being determined by taking into account pulses, called pulses of interest, formed by the detector during exposure to the source, or to a calibration source, during a time range of interest, the determination the correction factor comprising, for each pulse of interest, the following steps:
    i) first shaping of the pulse of interest, by considering the first shaping duration, and measurement of a first amplitude of the pulse of interest thus shaped;
    ii) second shaping of the pulse of interest, by considering a second shaping duration, different from the first shaping duration, and measurement of a second amplitude of the pulse of interest thus set in shape ;
    iii) comparing the first amplitude and the second amplitude of the pulse of interest, so as to calculate a comparison factor;
    the correction factor being determined as a function of the comparison factors calculated, during step iii), for each pulse of interest.
    A measurement pulse is thus obtained, the first amplitude of which is corrected.
    The second shaping time may be greater or less than the first shaping time.
    The process can also include:
    a second shaping of the measurement pulse, considering the second duration of shaping, and a determination of a second amplitude of the measurement pulse thus shaped;
    a comparison of the first amplitude and the second amplitude of the measurement pulse, so as to calculate, for the measurement pulse, a comparison factor; so that the correction factor is also determined as a function of the comparison factor calculated for the measurement pulse. The measurement pulse is then considered as an interest pulse.
    During step iii), the correction factor can be determined as a function of an average or a median value of the comparison factors respectively calculated for each pulse of interest, possibly taking into account the correction factor calculated for the measurement pulse.
    The time range of interest can extend below and / or beyond the measurement instant. It can understand the measurement instant or not understand the measurement instant.
    According to one embodiment, the time range of interest corresponds to a calibration phase, during which the detector is exposed to a calibration source, the method being such as during step iii), the determination of the factor for correction also includes, for each pulse of interest, the following steps:
    iv) third shaping of the pulse of interest, considering a third shaping duration, the third shaping duration being greater than the second shaping duration and the first shaping duration and determining a third amplitude of the pulse of interest thus formed;
    v) comparison of the third amplitude and the first amplitude of the pulse of interest, resulting from step i), or of the second amplitude of the pulse of interest resulting from step ii), so calculating a so-called auxiliary comparison factor;
    determining the correction factor also comprising the establishment of a correction function, representing the change, in particular the average change, in the comparison factor determined for pulses of interest as a function of the auxiliary comparison factor determined for said pulses pulses of interest, the method also comprising:
    a second shaping of the measurement pulse, considering the second shaping duration, and a determination of a second amplitude of the measurement pulse;
    a comparison of the first amplitude and the second amplitude of the measurement pulse, so as to calculate, for the measurement pulse, a comparison factor; so that during step c) the correction factor is determined as a function of a value of the correction function corresponding to the comparison factor calculated for the measurement pulse.
    According to one embodiment:
    steps a) and b) are implemented for a plurality of measurement pulses, during an acquisition period, the method comprising a step bj of forming a spectrum, representing a histogram of the first amplitudes formed during of each step b);
    steps i) to iii) are carried out for each measurement pulse acquired during the acquisition period, so that the time range of interest corresponds to the acquisition period;
    during step c), the method includes resetting the spectrum formed during step b '), taking into account the correction factor calculated during step iii), so as to form a corrected spectrum.
    Preferably, whatever the embodiment, the determination of the correction factor includes a step of selecting the pulses of interest, the step of selecting comprising, for each pulse of interest:
    a determination of a criterion of the pulse of interest or of the pulse resulting from the first or the second shaping of the pulse of interest;
    a comparison of the criterion with a threshold value;
    selecting the pulse of interest based on the comparison;
    so that steps i) to iii) are implemented only for the pulses of interest thus selected. Unselected pulses are then rejected.
    The selection step can include:
    determining a duration of the pulse of interest resulting from the first or the second shaping;
    a comparison of the determined duration with a threshold duration;
    a selection of the pulse of interest according to the comparison.
    The threshold duration can be a predetermined duration.
    The selection step can include:
    determining an area and an amplitude of the pulse resulting from the first or second shaping of the pulse of interest;
    calculating a ratio between the area and the amplitude thus determined;
    a comparison of the ratio with a threshold ratio value;
    a selection of the pulse of interest according to the comparison.
    The selection step may include a determination of the second time derivative of the pulse of interest or of the pulse resulting from the first or the second shaping of the pulse of interest, the pulse of interest being selected if the second time derivative does not cancel or change its sign.
    The selection step can include:
    determining a rise time of the pulse of interest or the pulse resulting from the first or the second shaping of the pulse of interest;
    a comparison between the determined rise time, with a threshold value of the rise time;
    - a selection of the pulse of interest according to the comparison.
    According to one embodiment, the detector comprises:
    a collection electrode, allowing the formation of the measurement pulse and the pulses of interest;
    an adjacent electrode of the collection electrode, the adjacent electrode being able to form a pulse, called adjacent pulse, the amplitude of which depends on an energy released by the ionizing radiation during its interaction in the detector.
    According to this embodiment, the determination of the correction factor may include a step of selecting the pulses of interest, the selection step comprising, for each pulse of interest:
    assigning a detection instant to the pulse of interest;
    an analysis of pulses formed by the adjacent electrode in a detection time range extending around the detection instant;
    a rejection of the pulse of interest when a pulse, formed by the adjacent electrode in the detection time range, exceeds an amplitude threshold
    According to one embodiment, during step b), the measurement pulse is shaped by applying a first time delay to the measurement pulse, to form a delayed measurement pulse, and by subtracting the measurement pulse delayed to the measurement pulse, the time delay corresponding to the first shaping time;
    during step i), the first shaping of the pulse of interest is carried out by applying the first time delay to the pulse of interest, to form a first delayed pulse of interest, and by subtracting the first interest pulse delayed to the interest pulse;
    during step ii), the second shaping of the pulse of interest is carried out by applying a second time delay to the pulse of interest, to form a second delayed pulse of interest, and by subtracting the second interest pulse delayed to the interest pulse, the second time delay corresponding to the second shaping duration.
    According to one embodiment, during step b), the measurement pulse is formed by applying a first filter to the measurement pulse, the first filter taking into account the first activation time. shape;
    during step i), the first shaping of the pulse of interest is carried out by applying the first filter to the measurement pulse;
    during step ii), the second shaping of the pulse of interest is carried out by applying the second filter to the pulse of interest, the second filter taking into account the second shaping duration.
    A second object of the invention is an electronic circuit for processing a pulse formed by a detector of ionizing radiation, the detector comprising a detector material, intended to interact with ionizing radiation, so as to form electrical charges during interaction of ionizing radiation in the detector;
    a preamplification circuit, able to collect the charges generated by the detector and to form a pulse, called measurement pulse, the amplitude of which depends on the quantity of charges collected;
    the processing circuit being characterized in that it comprises:
    a first shaping circuit, configured to shape the measurement pulse according to a first shaping duration, so as to generate a first shaped pulse;
    a second shaping circuit, configured to shape the measurement pulse according to a second shaping duration, greater than the first shaping duration, so as to generate a second shaping pulse;
    a comparison unit, able to compare the first shaped pulse and the second shaped pulse, so as to determine a comparison factor; a calculating unit, configured to determine a correction factor as a function of the comparison factor;
    a correction unit, capable of applying the correction factor determined by the calculation unit to the first pulse shaped so as to form a corrected pulse.
    According to one embodiment:
    the first shaping circuit comprises a first delay line, configured to apply a first delay, corresponding to the first shaping duration, to the measurement pulse, the first shaping circuit comprising a subtractor configured to subtract the pulse thus delayed from the measurement pulse;
    the second shaping circuit comprises a second delay line, configured to apply a second delay, corresponding to the second shaping duration, to the measurement pulse, the second shaping circuit comprising a subtractor configured to subtract the delayed pulse from the measurement pulse.
    According to one embodiment:
    the first shaping circuit is a filter able to generate, from the measurement pulse, a first pulse of predetermined shape parameterized by the first shaping duration;
    the second shaping circuit is a filter capable of generating, from the measurement pulse, a second pulse of predetermined shape parameterized by the second shaping duration.
    The calculation unit can be able to implement steps i) to iii) of the method according to the first object of the invention from pulses, called pulses of interest, formed by the detector during a time range d 'interest.
    Other advantages and characteristics will emerge more clearly from the description which follows of particular embodiments of the invention, given by way of nonlimiting examples, and represented in the figures listed below.
    FIGURES
    Figures IA and IB show an example of a detection device.
    Figure IC shows a preamplifier model.
    Figure 1D shows a model of a delay line shaping circuit.
    Figures 2A and 2B illustrate the operation of a delay line circuit. FIG. 2A shows an example of a measurement pulse, resulting from a preamplifier, of the measurement pulse delayed by a first time interval as well as of a pulse, called shaped pulse, obtained by subtracting the 'measurement pulse delayed to the measurement pulse. FIG. 2B shows an example of a measurement pulse, resulting from a preamplifier, of the measurement pulse delayed by a second time interval as well as a pulse, known as shaped pulse, obtained by subtracting the 'measurement pulse delayed to the measurement pulse.
    FIG. 2C illustrates the shaping of a measurement pulse by a trapezoidal filter. It represents a measurement pulse as well as two shaped pulses by considering respectively a first shaping duration and a second shaping duration. FIG. 3A shows two pulses shaped by considering a first shaping duration, these pulses being obtained at a first instant and at a second instant, the first and second instants being spaced two hours apart. FIG. 3B is similar to FIG. 3A, the pulses being shaped by considering a second shaping duration greater than the first shaping duration.
    FIG. 3C shows the time evolution of a spectrum of radiation produced by an isotopic source of 57 Co by implementing a shaping circuit based on a first shaping duration. Figure 3D shows the temporal evolution of a spectrum of radiation produced by an isotopic source of 57 Co by implementing a shaping circuit based on a second shaping time greater than the first shaping time in shape.
    FIG. 4A represents the main steps of an embodiment of the invention. FIG. 4B shows a circuit comprising two delay lines allowing two different shapings of a measurement pulse, respectively by considering a first shaping duration and a second shaping duration.
    FIG. 4C represents a processing circuit allowing the implementation of the method illustrated in connection with FIG. 4A.
    FIGS. 4D and 4E respectively show the application of a criterion for selecting a shaped pulse respectively according to a first shaping duration and a second shaping duration.
    FIG. 5A shows a temporal evolution of a correction factor.
    FIG. 5B illustrates two spectra of the radiation produced by an X-ray source, without implementing the invention, the respective acquisitions of the two spectra being offset by 120 minutes. FIG. 5C illustrates the radiation spectra described in connection with FIG. 5B after implementation of the invention.
    FIG. 6A represents the main steps of another embodiment of the invention.
    FIG. 6B shows a circuit comprising three delay lines allowing three different shapings of a measurement pulse, respectively by considering a first shaping duration, a second shaping duration and a third shaping duration . FIG. 6C represents the time evolution of a first comparison factor and of a second comparison factor.
    FIG. 6D shows an evolution of a so-called calibration comparison factor as a function of the first comparison factor represented in FIG. 6C.
    FIG. 6E illustrates a correction function, making it possible to establish a correction factor as a function of a comparison factor.
    EXPLANATION OF PARTICULAR EMBODIMENTS
    FIG. 1A shows a device 1 allowing the implementation of the invention. The device comprises a detector 10, capable of interacting with ionizing radiation 5 emitted by an irradiation source 2. By ionizing radiation is meant radiation formed by particles capable of ionizing the material. It can be alpha radiation, beta radiation, photon radiation of type X or gamma, or even neutron radiation. In the example shown, the radiation is photon radiation of type X or gamma, formed of photons whose energy is for example between 1 keV and 2 MeV.
    In the example shown, the detector comprises a semiconductor material, of CdTe type, but it could also be a semiconductor material commonly used for the detection of ionizing radiation, for example of the Ge type, Yes, CdZnTe. When a particle, in this case a photon, of ionizing radiation 5 interacts in the detector 10, charge carriers are formed and migrate towards a collection electrode 11, for example an anode. The quantity of charges Q collected by the electrode 11 depends, preferably linearly, on the energy E released by the particle under the effect of the interaction.
    Other types of detectors, for example scintillators coupled to a photon / charge carrier converter, or a gaseous detector of the ionization chamber type, can be used, provided that they allow the collection of a quantity of charges Q under the effect of an energy E released by the ionizing radiation during an interaction in the detector 10. In this description, the term amplitude designates the maximum height of a pulse. It can also be the integral of a pulse, or any other function of the maximum height or the integral of the pulse.
    The collection electrode 11 is connected to an electronic processing circuit 12, comprising:
    a preamplification circuit 20, for example a charge preamplifier, able to integrate the charges collected by the collecting electrode 11 and to form a pulse S, the amplitude A of which depends on the quantity of charges Q collected by the electrode 11 following an interaction of a particle in the detector 10. Such a preamplifier is shown diagrammatically in FIG.
    an amplifier 30, capable of amplifying and shaping the pulse S and forming a shorter pulse S ', so that it is less conducive to the formation of stacks. The pulse S'has an amplitude A '. An example of an amplifier is described below, in connection with Figure ID.
    a correction circuit 40, for correcting the shaped pulse generated by the shaping circuit 30 and forming a corrected pulse S c whose amplitude A ′ c , known as corrected amplitude, is such that A ′ c = - x A ', where η denotes a correction factor. An important aspect of the invention is the determination of the correction factor η, the latter being variable over time. This determination is described below.
    an analyzer 50 to analyze the corrected amplitude A ' c , and more precisely to classify it as a function of its value. It can be a single-channel or multi-channel analyzer, a multi-channel analyzer making it possible to form an Sp spectrum following the detection of a plurality of interactions. By energy spectrum is meant the amplitude distribution of pulses during the exposure of the detector to a radiation source. Such a spectrum is in the form of a histogram, each term Sp (i) of which represents the number of pulses detected, the amplitude of which is equal to i. By amplitude equal to i; an amplitude is understood to lie within an amplitude range Δί comprising the value i, and for example centered around the value i. The relationship between the amplitude of a pulse and the energy deposited during the interaction at the origin of the pulse, is generally linear, and defined by a gain, called conversion gain. Such a gain is defined during an energy calibration, during which the detector is subjected to radiation of known energy. Preferably, this gain should be as stable as possible over time.
    The electronic processing circuit 12 can comprise an analog-to-digital converter not shown in FIG. 1A, the latter being able to be disposed between the preamplifier 20 and the shaping circuit 30, or between the shaping circuit 30 and the shaping circuit correction 40, or between the correction circuit 40 and the analyzer 50, or downstream of the latter. Depending on the position of the analog to digital converter, the electronic circuits described above are either analog or digital.
    Figure IB shows a pixelated detector 10 comprising at least two collection electrodes 11 and lt, each collecting electrode being respectively connected to a processing circuit 12, 12 a. The collection electrode ll a is designated by the term adjacent electrode, because it is disposed adjacent to the collection electrode 11. The processing circuit 12 is such as that shown in connection with FIG. IA. The processing circuit 12 a comprises a preamplifier 20 a , an amplifier 30 a , a correction circuit 40 a and an analyzer 50 a such as those described in connection with the processing circuit 12. Usually, a pixelated detector can comprise several tens or even hundreds of collection electrodes. These electrodes can be arranged in a line or a two-dimensional matrix.
    FIG. 1C schematizes the preamplification circuit 20 previously described. It can be schematized by an amplifier 21 connected to a feedback loop RC, modeled by a resistor R in parallel with a capacitor C. The charges collected by a collection electrode, during an interaction, form a pulse, said measurement pulse, whose rise time t r fast and whose descent is slow, the descent can be modeled t according to an expression of the type e ~ Rc, where t denotes the time. The rise time t r corresponds to the time interval between the start of the measurement pulse S and its maximum. Such a curve is shown in Figures 2A and 2B. The rise time t r is usually designated by the term "rise time".
    FIG. 1D schematizes the amplifier 30, the latter comprising a delay line 31, applying a delay St to the pulse, so as to form a delayed pulse S st . A subtractor 32 makes it possible to form a shaped pulse S ', so that S' = S - S st - The principle of such a delay line is known, and is described more precisely in patent US7652242. The shaped pulse S 'is more symmetrical than the pulse S coming from the preamplifier. Its amplitude A ′ depends, preferably by a linear relationship, on the energy transferred by the ionizing radiation during its interaction in the detector 10. FIGS. 2A and 2B represent two situations, corresponding respectively to a first delay line applying a first delay St x and to a second delay line applying a second delay 5t 2 θ the measurement pulse S, with St 2 > St 1 . In this example, St 2 > t r > St ^. The first delay line performs so-called short formatting, while the second delay line performs long formatting. In these figures, a pulse S has been shown coming from a preamplifier, the delayed pulses and Sg t2 as well as the pulses and S ' 2 coming from the shaping circuit respectively according to the first delay δί ± and the second delay St 2 , so that :
    S'i = S - S sti and S ' 2 = S - S stz
    The rise time t ' r of a shaped pulse, corresponding to the start of the shaped pulse and its maximum, usually designated by the term peaking time, is frequently greater than or equal to the rise time t r of l measurement pulse S. In this case, the amplitude of the shaped pulse takes into account the totality of the charge Q collected by the electrode. Otherwise, as can be seen in FIG. 2A, when the rise time t ' r of the shaped pulse S' is less than the rise time t r of the measurement pulse S, a fraction of the charge Q is not taken into account in the amplitude of the shaped pulse, this fraction not taken into account being designated by the term ballistic deficit. The peaking time t ' r depends on the shaping time St taken into account by the shaping circuit 30 of the measurement pulse.
    As mentioned in connection with the prior art, when the intensity of the incident ionizing radiation 5 is high, the flow of particles reaching the detector is high and the number of pulses S formed per unit of time, designated by the term rate of counting, increases. This results in the appearance of stacks, a stack corresponding to a configuration during which two interactions, coming from two different particles, are sufficiently temporally close together to be forming the same pulse, the amplitude of which corresponds to the cumulative amount of charge collected. following each interaction. Thus, the pulse does not carry relevant information as to the energy of the incident radiation. Such a phenomenon can for example occur in baggage screening type applications, in which baggage is exposed to intense radiation beams, so as to minimize the screening time. It can also occur when carrying out nuclear measurements under strong radiation, for example on high activity nuclear components. In applications with a high counting rate, too high a peaking time t ' r increases the probability of detection of stacks. It is necessary to reduce it, and to accept a certain ballistic deficit. However, it has been demonstrated that due to fluctuations appearing in the polarization of the detector material, when the peaking time t ' r is less than the rise time t r of the measurement pulse supplied by the preamplifier 20, l' amplitude of the shaped pulse S 'varies over time This variation is shown diagrammatically in FIG. 3A, representing pulses S' x shaped according to a first delay δί ± less than the rise time of the measurement pulse S , from the preamplifier.
    FIG. 3A shows a first pulse and a second pulse S '^ t ^ formed respectively from a measurement pulse 5 (^) at a first instant t ± and from a measurement pulse S (t 2 ) at a second instant t 2 . The two pulses S'jXG) and result from a delay line shaping circuit, taking into account a delay δί ± of 50 ns. The measurement pulses S (tj) and Sftj) processed by the shaping circuit have a rise time t r of 100 ns. They result from the exposure of a CdTe type detector of thickness 3 mm exposed to an isotopic source of 57 Co. The first shaped pulse was obtained at an instant t r close to the commissioning of the detector. The second pulse was obtained at a time t 2 posterior to 120 minutes at time t r . The pulses S'and S'i (t 2 ) correspond to the emission peak of 57 Co, at the energy 122 keV. We note that their respective amplitudes A '^ t ^ and A' - ^ Çt ^ are not identical, whereas they correspond to the same energy. Thus, in such a configuration, based on a short shaping time δί ± , and less than the rise time t r of the measurement pulse S, the conversion gain, corresponding to a ratio between the amplitude A of the shaped pulse S and the energy of the interaction which generated the measurement pulse S, is not stable over time.
    FIG. 3B represents two pulses and S ′ 2 (t 2 ) shaped in the same experimental conditions as the pulses of FIG. 3A, by modifying the shaping time. In the case of FIG. 3B, a delay line was used based on a second delay 5t 2 equal to 200 ns. It can be seen that their respective amplitudes A ′ 2 (tf) and Ai 2 (t 2 ) are identical, which attests to the stability of the conversion gain in this configuration based on a long shaping time 5t 2 .
    The drift effect of the conversion gain as a function of the shaping time can be observed in FIGS. 3C and 3D, representing energy spectra of the radiation emitted by a source of 57 Co acquired at different times, respectively by adopting a short (50 ns) and long (200 ns) shaping time. In FIG. 3C, the drift of the conversion gain results in a progressive shift of the spectra towards the high energies. In Figure 3D, the different spectra are superimposed, the conversion gain being stable.
    It should be noted that this effect does not depend on the type of shaping circuit used. The examples illustrated in FIGS. 3A to 3D were obtained by a delay line type shaping circuit. The same effect can be observed on other types of shaping circuits, based on a variable shaping time. For example, the shaping circuit can be a filter making it possible to generate a pulse of configurable shape, for example a Gaussian, triangular or trapezoidal shape, this type of circuit being usual in the field of X or gamma spectrometry. The impulse response of such a circuit is an impulse of predetermined geometric shape and configurable by the duration of shaping. The use of Gaussian filters, generating a Gaussian-shaped pulse and whose width at half-height is configurable, is common in the field of gamma or X spectrometry. The use of such filters is described in the publication Salathe M. Optimized digital filtering techniques for radiation detection with HPGe detectors. The filter can also be triangular, the base of the triangle being configurable, or trapezoidal, the width of the top of the pulse being configurable. FIG. 2C schematizes for example a pulse S at the output of a preamplifier, as well as pulses S '^ and S 2 shaped with a trapezoidal filter by considering respectively a first shaping duration St 1 and a second duration 5t 2 for shaping, with 5t 2 ><5 ^.
    Thus, a short formatting time δί ± limits stacking, but is accompanied by a temporal drift of the conversion gain. A long 5t 2 shaping time increases the probability of observing a stack, but benefits from a stable conversion gain. The inventors have used this observation to propose a method, the main steps of which are described below, in connection with FIGS. 4A and 4C.
    Step 100: detection: the detector 10 is exposed to incident ionizing radiation 5, a particle of which interacts in the detector to form charge carriers collected by an electrode.
    Step 110: preamplification: the collected charge Q is integrated by a preamplifier, so as to form a pulse S, called measurement pulse, generally asymmetrical, having a rise time t r and an exponential decrease. The measurement pulse is formed at a measurement instant t.
    Step 130: first shaping of the measurement pulse S according to a first shaping duration δί 1 . In this example, the shaping is obtained with a first delay line circuit 31, as shown in FIG. 4B, the delay line generating a delayed pulse S ôti , as explained in connection with FIG. 2A. The first delay line 31 is connected to the inverting input of a subtractor 34, the non-inverting input of which receives the measurement pulse S. The first shaping time St 1 is less than the rise time t r of the measurement pulse S at the output of the preamplifier 20. This step generates a first shaped pulse S ' ± of first amplitude A .
    Step 140: second shaping of the measurement pulse S according to a second shaping time ot 2 . In this example, the shaping is obtained with a second delay line circuit 32 as shown in FIG. 4B, the delay line generating a delayed pulse S st2 , as explained in connection with FIG. 2B. The second delay line 32 is connected to the inverting input of a subtractor 35, the non-inverting input of which receives the measurement pulse S, coming from the preamplifier 20. The second shaping time 8t 2 can be greater or equal to the rise time t r of the output pulse S of the preamplifier 20. This step generates a second shaped pulse S ' 2 of second amplitude A' 2 .
    The steps 150, 160 and 170 described below are implemented by the correction circuit 40, illustrated in FIG. 4C. The correction circuit 40 includes a comparison unit 44, a calculation unit 46 and a correction unit 48. It can include a selection unit 42 when the optional step 120, described below, is implemented. The use of selection step 120 is however recommended.
    Step 150: comparison of the first and second amplitudes. During this step, the comparison unit 44 performs a comparison of the amplitudes A and A ' 2 of the respective pulses S and S' 2 , so as to calculate a comparison factor k. In this example, the comparison factor k is a ratio between the amplitudes A and A ' 2 , for example k =
    The objective of the following steps is to multiply the first amplitude measured A by a correction factor -, so as to compensate for the time drift of the conversion gain. The correction factor r can be equal to the comparison factor k. However, the inventors considered that it was preferable to establish a correction factor from correction factors k calculated respectively for a certain number of pulses, called pulses of interest.
    Step 160: calculation of the correction factor. From the comparison factor calculated during step 150, the calculation unit 46 determines a correction factor η. The correction factor η is determined as a function of comparison factors k ± , ... k N respectively obtained by applying steps 110 to 150 to pulses called pulses of interest, formed before and / or after the pulse of measurement S. In other words, if t designates the measurement instant, at which the measurement pulse S is formed, the pulses of interest are formed by the detector before or after the measurement instant t. The term pulse of interest designates a pulse generated by the detector, before or after the measurement pulse S, and the comparison factor of which is kept. The correction factor η can be calculated as a function of the mean or median value of the comparison factors determined for different pulses of interest prior to the measurement pulse, and stored in the calculation unit 46, to which one add the comparison factor k established for the measurement pulse S, calculated during step 150. In this example, η = mean (k 1 , ...., k N , k). The pulses of interest S 1 .... S N are the N pulses extending during a time range Δί called the time range of interest, this time range extending posteriorly and / or prior to the instant measurement t at which the measurement pulse S is formed. The number N is an integer designating the number of pulses of interest used to determine the correction factor. It can be between 1 and 10,000, or even more. The application of the correction factor η to the amplitude of the pulse S shaped according to a short time interval St 1 makes it possible to correct the time drift of the conversion gain of the detector, as described in connection with FIGS. 3A and 3D. The higher N, the more stable the average of the comparison factors and less sensitive to detection noise. Too high a number N can affect the accuracy of monitoring the evolution of the detector gain. The number N can be determined by a person skilled in the art, as a function of experimental tests making it possible to assess the evolution of the conversion gain. According to this embodiment, the time range of interest Δί extends before, around, or after the measurement instant t, the duration of the time range of interest can for example be 1 second or a few seconds . In general, the duration of the time range of interest Δί is defined so that it can be considered that during this range, the conversion gain of the detector does not change, or negligibly.
    According to one embodiment, after the detector 10 has been put into service, during the first pulse, no comparison factor has been measured. In this case, N = 0. The first correction factor is only determined as a function of the comparison factor calculated for the measurement pulse. Then, as each measurement pulse S is processed, the number N is incremented at each measurement pulse, until a predetermined maximum value N max is reached. The correction factor η is then determined by carrying out a sliding average (or median) of the N max comparison factors respectively determined for the N max pulses of interest prior to the measurement pulse.
    According to a variant, after the detector has been put into service, the number N of pulses of interest whose comparison factor is taken into account for the calculation of the correction factor is fixed at a predetermined value. No correction factor is applied to the measurement pulses, until there are N pulses of interest whose comparison factor is determined. From this instant, the correction factor is determined by carrying out a sliding average of the N comparison factors respectively determined for the N pulses of interest.
    The pulses of interest can be distributed according to a temporal range of interest extending before and after the measurement pulse S. It can for example be Nm “ x pulses acquired before the measurement pulse and Nm “ X pulses acquired after the measurement pulse. In this case, steps 160 and 170 are implemented in post-processing, that is to say after the time range of interest Δί. The correction is then deferred with respect to the measurement instant t.
    According to another variant, the correction factor η is determined without taking into account the comparison factor of the measurement pulse, but only on the basis of the comparison factors respectively determined from pulses of interest.
    Step 170: correction. The correction factor η is applied to the first shaped pulse S ' it whose amplitude A is multiplied by -. We then obtain a corrected amplitude A ' c = - A . The correction is carried out by a correction unit 48 of the processing circuit
    40. The correction unit uses the correction factor η calculated by the calculation unit 46.
    At the end of step 170, steps 100 to 170 are repeated upon detection of another measurement pulse S.
    According to a variant, each comparison factor comprises a normalized difference between the amplitudes A ′ 2 and Af. Thus, each comparison factor k is such that k = An average of the comparison factors is carried out, so as to obtain a correction factor η, the corrected amplitude A ' c being such that A' c = A + η A = A'fl + η).
    Thus, the method described in connection with steps 100 to 170 makes it possible to determine an amplitude A of pulses S shaped according to a short shaping time, which is compatible with use of the detector in high rate of counting. The impulse thus shaped is corrected, so as to attenuate, or even eliminate the effect of drift of the conversion gain highlighted by the inventors. The advantage is to use a short shaping time, favorable to high counting rates, while limiting the drawback linked to the temporal instability of the conversion gain.
    According to a preferred embodiment, the method comprises a step 120 of selecting the pulses for which the comparison factors k are determined: these are the pulses of interest and possibly the measurement pulse. Step 120 is implemented by the selection unit 42. The objective is to avoid taking into account comparison factors originating from pulses formed following a stack of two interactions originating from two different particles, or of pulses resulting from a collection of charge carriers shared between two adjacent electrodes, when the detector is connected to several electrodes as shown in FIG. 1B. The selection makes it possible to retain only pulses formed by a single particle, and collected by a single electrode. The selection is described below in connection with a measurement pulse, knowing that it can be carried out for each pulse of interest.
    To detect the presence of a stack, the selection unit 42 applies one or more selection criteria to the measurement pulse S from the preamplifier 20, or to the first or the second shaped pulse. A selection criterion may be a duration of the first pulse shaped S and / or or of the second pulse shaped S ' 2 above a predefined threshold. This selection criterion is illustrated in Figures 4D and 4E. In FIG. 4D, a pulse S shaped by the first shaping circuit has been represented, considering the first short time interval δί ± of 50 ns. The duration of the pulse above an amplitude threshold equal to 0.2 is indicated by a double arrow. When the duration At 1 exceeds a maximum duration, for example 80 ns, it is considered that the pulse results from a stacking and the pulse is rejected. In FIG. 4E, a pulse S ' 2 has been shown shaped by the second shaping circuit, considering the second time interval ot 2 of 200 ns. The duration of the pulse Δί 2 above an amplitude threshold equal to 0.2 is shown by a double arrow. When the duration Δί 2 exceeds a maximum duration, for example 230 ns, it is considered that the pulse results from a stack and the pulse is rejected.
    Another selection criterion can be based on the rise time t r of a measurement pulse S, or on the rise time (peaking time) t ' rl of a pulse S formed by the first circuit of shaping, or on the peaking time t ' r 2 of a pulse S' 2 shaped by the second shaping circuit. When at least one of these rise times exceeds a certain value, the corresponding pulse is rejected.
    When the detector is connected to several collecting electrodes, as shown in FIG IB, a selection criterion is a coincident detection of a pulse on an adjacent electrode ll. Upon interaction, the generated charge Q can be collected by an electrode 10 while a further charge is collected by an adjacent lt electrode has, in a step 100 a. We are then in the presence of a charge sharing, the charge collected by each electrode not being representative of the energy deposited in the detector during the interaction. The additional charge Q a is integrated by a preamplifier 20 a during a step 120 a , so as to form an adjacent pulse S a . The adjacent pulse S a is shaped by the adjacent amplifier 30 a so as to obtain a first adjacent pulse shaped S ' al according to the first shaping time Sty (step 130 a ) and a second adjacent pulse S ' a2 shaping according to the second shaping time ôt 2 (step 140 a ). The selection unit 42 detects the presence of the adjacent pulse S a and / or the first adjacent shaped pulse S ' la and / or the second adjacent shaped pulse S' 2 a . When at least one of these pulses occurs simultaneously with the measurement pulse S coming from the collection electrode 10, or with the first shaped pulse S , or with the second shaped pulse S ' 2 , we consider that a load sharing has taken place and the measurement pulse S is rejected. For this, the selection unit compares the amplitudes A ′ la and A ′ 2a of the pulses S ′ la and S ′ 2 a to a predetermined threshold. When an amplitude exceeds the threshold, there is a coincident detection. By coincidence is meant occurring at two sufficiently close times, for example in a time range known as coincidence T d of 10 or 20 ns around the detection instant t d of a pulse of interest.
    The selection unit 42 can also be based on other selection criteria, for example: the presence of an inflection point on the measurement pulse or on a shaped pulse S or S ' 2 , such presence may indicate stacking.
    A ratio between the amplitude (Af or A ' 2 ) of a shaped pulse (5 or S' 2 ) over the integral of said pulse. Too high a ratio may indicate the presence of a stack.
    An amplitude of the measurement pulse S or of a shaped pulse (S ^ or S ' 2 ) below a predetermined threshold value, below which the pulse is considered to be non-significant, and caused by noise .
    Too high a count rate, the count rate representing the number of pulses detected per unit of time.
    Thus, the selection unit 42 is based on one or more selection criteria to reject pulses which are not representative of the charge deposited by an interaction of a particle in the detector medium 10. These undesirable pulses are not taken into account in the calculation of the correction factor k. The selection step 120 makes it possible to obtain a correction factor independent of the counting rate of the detector. The pulses selected by the selection unit 42 are processed by the comparison unit 44, the calculation unit 46 and the correction unit 48.
    The previously described experimental device was subjected to prolonged irradiation, the irradiation source being an X-ray generator brought to a voltage of 120 kV. FIG. 5A represents the evolution of the correction factor η as a function of time, from the powering up of the detector (t = 0) until 2 hours after this powering up (t = 120 minutes). The correction factor was calculated by taking a sliding average of 1024 comparison factors. It is observed that the correction factor η varies between an initial value of approximately 0.785 to a final value close to 0.835. This curve shows the relevance of considering a correction factor that varies over time.
    FIG. 5B shows spectra of the radiation 5 emitted by the X-ray generator and detected by the detector 10 respectively at t = 2 min and at t = 120 min after the commissioning of a detector 10, with shaping of the pulses with the first shaping circuit, based on a shaping time of 50 ns, without pulse correction. We observe a spectral drift towards high energies. FIG. 5C shows spectra of the radiation 5 detected by the detector 10 respectively at t = 2 min and at t = 120 min, with implementation of steps 100 to 170, that is to say by correcting each pulse set formed by the first shaping circuit, by applying the correction factor shown in FIG. 5A. It is observed that the two spectra are not offset from each other, but overlap, which attests to the reliability of the correction.
    FIG. 6A represents another embodiment of a method according to the invention. Steps 100, 110,120, 100 α , 110 α , 130 a and 140 a are identical to those described in connection with FIG. 4A. According to this embodiment, the first shaping time St 1 and the second shaping time δί 2 are less than the rise time t r of the measurement pulse S. For example, δί ± = 50 ns and δί 2 = 30 ns. During step 150, the comparison circuit 44 determines a comparison factor k by comparison between the amplitudes A and AÏ 2 , for example k = It is noted that in this example, ôt 2 > δί ±
    During step 160, from the comparison factor k established for the measurement pulse, the correction circuit 46 determines a correction factor η such that η = f (k). The function f is obtained during a preliminary calibration phase, described in connection with steps 90 to 98.
    During step 170, the correction unit applies the correction factor η to the first shaped pulse S ' lt such that A' c = - Α'γ.
    A first difference between this embodiment and the previous embodiment, represented in FIG. 4A, is that the shaping circuits are based on shaping times shorter than the rise time t r of the measurement pulse S from the preamplifier 20. This makes it possible to limit stacking and this embodiment is therefore well suited to high counting rates.
    A second difference is that this embodiment presupposes a prior calibration phase, described in connection with steps 80 to 98. This calibration phase corresponds to a time range of interest Δί, during which pulses of interest are acquired in order to have comparison factors. In the previous embodiment, the pulses of interest are measurement pulses formed during the same measurement sequence, before and / or after each measurement pulse, using the same irradiation source. According to the present embodiment, the pulses of interest are formed during a prior calibration phase. This calibration phase is preferably implemented using a calibration source which does not expose the detector to too intense incident radiation, so that the counting rate is low, which limits the probability of detection of stacks. . During the calibration phase, the amplifier 30 comprises three shaping circuits, as described in FIG. 6B:
    a first shaping circuit, based on a first shaping duration St 1 , and preferably δί ± <t r. In this example, ôty = 50 ns.
    a second shaping circuit, based on a second shaping duration δί 2 with δί 2 Φ δί ± and, preferably, δί 2 <t r . In this example, 5t 2 = 30 ns.
    a third shaping circuit, based on a third shaping duration 5t 3 with 5t 3 > δ1 2 , δί 3 > δί 1 , preferably, δί 3 > t r . In this example, δί 3 = 200 ns.
    In the example shown in FIG. 6B, the first, second and third shaping circuits are respectively delay line circuits comprising respectively the first delay line 31 and the second delay line 32 described in connection with the figures IC and
    1D, as well as a third delay line 33 applying a time delay equal to ôt 3 to the pulse S coming from the preamplifier 20. The third shaping circuit is connected to the inverting input of a subtractor 36, of which the non-inverting input receives the pulse S resulting from the preamplifier 20.
    The calibration step is now described, in connection with steps 80 to 98.
    Step 80: detection: the detector is exposed to incident ionizing radiation, a particle of which interacts in the detector to form charge carriers Q collected by an electrode.
    Step 90: preamplification: the charge collected is integrated into a preamplifier, so as to form a calibration pulse S ca i ib .
    Step 91: first shaping of the calibration pulse S ca i ib generated by the preamplifier, according to the first shaping duration St 1 , so as to obtain a first amplitude shaped in S ' caÎib _ 1 of first amplitude A ' ca i ib _ 1 .
    Step 92: second shaping of the calibration pulse S ca i ib generated by the preamplifier, according to the second shaping duration ot 2 so as to obtain a second amplitude shaped S ' ca i ib _ 2 of second amplitude A ' ca i ib _ 2 .
    Step 93: third shaping of the calibration pulse S ca i ib generated by the preamplifier, according to the third shaping duration St 3 so as to obtain a third amplitude shaped S ' ca i ib _ 3 of third amplitude A ' ca i ib _ 3 .
    Step 94: calculation of a first comparison factor k ± by comparison between the first amplitude A ' caÎÎb _ 1 and the third amplitude A' ca i ib _ 3 . For example, k r = A ' callb ~ The first A ' calib-3 comparison factor k ± is hereinafter called the auxiliary comparison factor.
    Step 95: calculation of a second comparison factor k 2 by comparison between the second amplitude A ' caLÎb _ 2 and the third amplitude A' ca i ib _ 3 . For example, k 2 = A ' callb ~ 2 _ A ' calib-3
    Step 96: repetition of steps 90 to 95, so as to establish a succession of couples (k lt k 2 ) as a function of time. FIG. 6C represents a temporal evolution, over a duration of 120 minutes, of the first comparison factor k ± and of the second comparison factor / c 2 24
    Step 98: development of the correction function. It is a question of establishing a correction function kf, making it possible to establish a relationship between the ratio - and the first correction factor, known as ^ 2 Ar calib-1 auxiliary correction factor k,. We notice that - = ^, cal f ~ 3 = A ' caÎlb ~ 1 , The ratio - ^ 2 —cahb — 1 ^ f calib-2 ^ 2 A ' calib-3 corresponds to an amplitude shaped by the first circuit shaping on an amplitude shaped by the second shaping circuit. It therefore corresponds to the comparison factor k previously described, except that it is established for the calibration pulses. It is noted k ca i ib .
    FIG. 6D, obtained from the data in FIG. 6C, makes it possible to establish a temporal evolution k of the ratio k ca i ib = -. However, the first comparison factor k ± corresponds to the fc 2 correction factor η, allowing correcting each first amplitude A so as to obtain a corrected amplitude A c , in order to estimate an amplitude shaped with a long shaping time, that is to say greater than or equal to the rise time t r . Furthermore, the temporal evolution of k r is known (cf. FIG. 6C). By crossing the data in FIGS. 6C and 6D, a correction function f can be obtained by determining the evolution, over several calibration pulses, of an average (or of a median) of auxiliary correction factors k r as a function an average (or median) of calibration factors k ca i ib . The correction function f represents the evolution of the correction factor η as a function of the comparison factor k. This function is shown in Figure 6E. It results from the calibration phase.
    From the correction function f, knowing the comparison factor k of a measurement pulse, the correction factor η is obtained according to the expression η = f (k). The correction is then carried out so that the corrected amplitude A ' c is such that A' c = - Af.
    According to a variant, the comparison factor k calculated during step 150 is such that k =
    In this case, for each calibration pulse, k ca i ib = is determined = During calibration, the evolution of the second comparison factor k 2 is determined as a function of ratio k ca i ib , so as to obtain the function correction f. The second comparison factor then corresponds to the auxiliary correction factor. During step 160, the correction factor η = f (k calib ) is determined and during step 170, the correction is carried out by applying the correction factor η to the second corrected amplitude A ′ 2 so that A ' c = - A' 2 .
    Generally, the first comparison factor k 1 or the second comparison factor k 2 are designated by the term auxiliary comparison factor.
    According to one embodiment, the method comprises the acquisition of measurement pulses S, during an acquisition period T a , the measurement pulses S processed by the first shaping circuit being classified according to their amplitudes A respective, so as to form a spectrum Sp. The spectrum Sp corresponds to a histogram of the amplitudes of the pulses processed by the first shaping circuit. Furthermore, each measurement pulse S is also processed by the second shaping circuit, so that there is a comparison factor k for each measurement pulse. According to this embodiment, the time range of interest At corresponds to the acquisition period T a . A correction factor η is defined for the time range of interest At from the comparison factors. The correction is applied not successively, to each amplitude acquired, but simultaneously, by correcting the spectrum Sp, so that:
    Sp c (i ') = Sp (j] xj) xn, where:
    Sp c is the corrected spectrum;
    i is a corrected spectrum channel;
    η is the correction factor, which is determined as a function of the correction factors respectively calculated for the pulses composing the spectrum;
    n is a normalization term, defined so that the integral of the corrected spectrum Sp c corresponds to the integral of the uncorrected spectrum Sp. For example, n = i is a channel of the spectrum Sp before the correction.
    The correction amounts to resetting the spectrum Sp in the direction of the weaker channels, so as to correct the spectral drift resulting from the progressive increase in the conversion gain.
    This embodiment assumes that the conversion gain is stable during the acquisition period T a of the spectrum Sp. This embodiment makes it possible to simultaneously correct the amplitudes A shaped by the first shaping circuit.
    Furthermore, as previously mentioned, whatever the embodiment, the method may include a digitization step, for example between the preamplifier and the amplifier, either between the amplifier and the correction circuit, or downstream of this latest.
    The invention may be implemented in measurement applications with a high counting rate, the applications possibly being in non-destructive testing, baggage screening, nuclear power or the field of diagnosis or monitoring of medical treatment.
    权利要求:
    Claims (18)
    [1" id="c-fr-0001]
    1. Method for processing a pulse (S) generated by a detector (10) of ionizing radiation, the detector (10) being able to interact with ionizing radiation (5) to generate said pulse, the amplitude (ri ) depends on an energy released by ionizing radiation during its interaction in the detector, the process comprising the following steps:
    a) exposure of the detector to a source of ionizing radiation (2) so as to obtain, at a measurement instant (t), a pulse (S), said measurement pulse;
    b) shaping the measurement pulse, considering a first shaping duration (5 ^), and determining a first amplitude (ri ' x ) of the measurement pulse thus shaped;
    c) correction of the first amplitude determined during step b), taking into account a correction factor (η), so as to obtain a corrected amplitude (ri ' c );
    the correction factor (η) being determined by taking into account pulses, called pulses of interest (S x , ... S N , S ca i ib ), formed by the detector during exposure to the source or to a calibration source, during a time range of interest (Δί), the determination of the correction factor comprising, for each pulse of interest, the following steps:
    i) first shaping of the pulse of interest, considering the first shaping duration (5 ^), and measurement of a first amplitude of the pulse of interest thus shaped;
    ii) second shaping of the pulse of interest, by considering a second shaping duration (ôt 2 ), different from the first shaping duration, and measurement of a second amplitude of the pulse d interest thus shaped;
    iii) comparison of the first amplitude and the second amplitude of the pulse of interest, so as to calculate a comparison factor (k, k lt ...., k N , k ca i ib );
    the correction factor (η) being determined as a function of the comparison factors (k, k ± , ...., k N , k ca i ib ) calculated, during step iii), for each pulse of interest .
    [2" id="c-fr-0002]
    2. Method according to any one of the preceding claims, also comprising:
    a second shaping of the measurement pulse, by considering the second shaping duration (5t 2 ), and a determination of a second amplitude (ri ' 2 ) of the measurement pulse thus shaped;
    a comparison of the first amplitude and the second amplitude of the measurement pulse, so as to calculate, for the measurement pulse, a comparison factor (k) ·, so that the correction factor (η) is also determined based on the comparison factor (k) calculated for the measurement pulse.
    [3" id="c-fr-0003]
    3. Method according to any one of the preceding claims, in which during step iii), the correction factor (η) is determined as a function of an average value or of a median value of the comparison factors respectively calculated. for each pulse of interest.
    [4" id="c-fr-0004]
    4. Method according to any one of the preceding claims, in which the time range of interest (Δί) extends below and / or beyond the measurement instant (t).
    [5" id="c-fr-0005]
    5. Method according to any one of claims 1 to 3, in which the time range of interest (Δί) corresponds to a calibration phase, during which the detector is exposed to a calibration source, the method being such that during step iii), the determination of the correction factor also comprises, for each pulse of interest, the following steps:
    iv) third shaping of the pulse of interest, considering a third shaping duration (5i 3 ), the third shaping duration being strictly greater than the second shaping duration (5i 2 ) and at the first shaping time (5 ^), and determination of a third amplitude (Τ1 ' εαίίϋ _ 3 ) of the pulse of interest thus shaped;
    v) comparison of the third amplitude with the first amplitude of the pulse of interest resulting from step i) or with the second amplitude of the pulse of interest, resulting from step ii), so as to calculate a so-called auxiliary comparison factor (ki, k 2 );
    the determination of the correction factor also comprising the establishment of a correction function (f), representing the evolution of the comparison factor (/ c ca H /,) determined for pulses of interest (S caUb ) as a function the auxiliary comparison factor determined for said pulses of interest, the method also comprising:
    a second shaping of the measurement pulse, considering the second shaping duration (5i 2 ), and a determination of a second amplitude of the measurement pulse (A ' 2 );
    a comparison of the first amplitude (4 ^) and the second amplitude (A ' 2 ) of the measurement pulse, so as to calculate, for the measurement pulse, a comparison factor (k);
    so that during step c) the correction factor (η) is determined as a function of a value of the correction function (/) corresponding to the comparison factor (k) calculated for the measurement pulse.
    [6" id="c-fr-0006]
    6. Method according to any one of claims 1 to 4, in which:
    steps a) and b) are implemented for a plurality of measurement pulses, during an acquisition period (T a ), the method comprising a step b ') of forming a spectrum (Sp), representing a histogram of the first amplitudes shaped during each step b);
    steps i) to iii) are carried out for each pulse acquired during the acquisition period (T a ), so that the time range of interest (Δί) corresponds to the acquisition period;
    during step c), the method comprises a registration of the spectrum (Sp) formed during step b '), taking into account the correction factor (η) calculated during step iii), so to form a corrected spectrum (Sp c ).
    [7" id="c-fr-0007]
    7. Method according to any one of the preceding claims, in which the determination of the correction factor comprises a step of selecting the pulses of interest, the step of selecting comprising, for each pulse of interest:
    a determination of a criterion of the pulse of interest or of the pulse resulting from the first or the second shaping of the pulse of interest;
    a comparison of the criterion with a threshold value;
    selecting the pulse of interest based on the comparison;
    so that steps i) to iii) are implemented only for the pulses of interest thus selected.
    [8" id="c-fr-0008]
    8. Method according to claim 7, in which the selection step comprises:
    determining a duration of the pulse of interest resulting from the first or the second shaping;
    a comparison of the determined duration with a threshold duration;
    a selection of the pulse of interest according to the comparison.
    [9" id="c-fr-0009]
    9. The method of claim 7, wherein the selection step comprises: determining an area and an amplitude of the pulse resulting from the first or the second shaping of the pulse of interest ;
    calculating a ratio between the area and the amplitude thus determined;
    a comparison of the ratio with a threshold ratio value;
    a selection of the pulse of interest according to the comparison.
    [10" id="c-fr-0010]
    10. The method as claimed in claim 7, in which the selection step comprises a determination of the second time derivative of the pulse of interest or of the pulse resulting from the first or from the second shaping of the pulse. of interest, the interest pulse being selected if the second time derivative does not cancel or change its sign.
    [11" id="c-fr-0011]
    11. The method as claimed in claim 7, in which the selection step comprises:
    determining a rise time of the pulse of interest or the pulse resulting from the first or the second shaping of the pulse of interest;
    a comparison between the determined rise time, with a threshold value of the rise time;
    a selection of the pulse of interest according to the comparison.
    [12" id="c-fr-0012]
    12. Method according to any one of the preceding claims, in which the detector (10) comprises:
    a collection electrode (11), allowing the formation of the measurement pulse and the pulses of interest;
    an adjacent electrode (ll a ) of the collecting electrode, the adjacent electrode being able to form a pulse, called adjacent pulse, the amplitude of which depends on an energy released by the ionizing radiation during its interaction in the detector (10);
    and in which the determination of the correction factor (k) comprises a step of selecting the pulses of interest, the selection step comprising, for each pulse of interest: an assignment of a detection instant (t d ) to the impetus of interest;
    an analysis of pulses (S a ) formed by the adjacent electrode (ll a ) in a coincidence time range (T d j extending around the time of detection;
    a rejection of the pulse of interest when a pulse (S a ), resulting from the adjacent electrode (1 l a ), in the time range of coincidence, exceeds an amplitude threshold.
    [13" id="c-fr-0013]
    13. Method according to any one of the preceding claims, in which: during step b), the shaping of the measurement pulse (S) is carried out by applying a first time delay to the measurement pulse , to form a delayed measurement pulse, and by subtracting the delayed measurement pulse from the measurement pulse, the time delay corresponding to the first shaping time (St /;
    during step i), the first shaping of the pulse of interest (S x .... S N , S caÎib ) is carried out by applying the first time delay to the pulse of interest, for forming a first delayed interest pulse, and subtracting the first delayed interest pulse from the interest pulse;
    during step ii), the second shaping of the pulse of interest (S x ... -S N , S ca i ib ) is carried out by applying a second time delay to the pulse of interest , to form a second delayed interest pulse, and by subtracting the second delayed interest pulse from the interest pulse, the second time delay corresponding to the second shaping duration (St /.
    [14" id="c-fr-0014]
    14. Method according to any one of claims 1 to 12, in which:
    during step b), the measurement pulse (S) is shaped by applying a first filter to the measurement pulse, the first filter taking into account the first shaping time (St /;
    during step i), the first shaping of the pulse of interest (S x .... S N , S caÎib ) is carried out by applying the first filter to the measurement pulse;
    in step ii), the second arrangement of the pulse of interest (S x ... .S N, Sc has ub) e st performed by applying the second filter to the pulse of interest, second filter taking into account the second shaping time (St /.
    [15" id="c-fr-0015]
    15. Electronic circuit for processing (12) a pulse formed by an ionizing radiation detector, the detector comprising a detector material (10), intended to interact with ionizing radiation, so as to form electric charges during 'an interaction of ionizing radiation in the detector;
    a preamplification circuit (20), able to collect the charges generated by the detector and to form a pulse (S), called measurement pulse, the amplitude of which depends on the quantity of charges collected;
    the electronic processing circuit being characterized in that it comprises:
    a first shaping circuit (31,34), configured to shape the measurement pulse according to a first shaping duration so as to generate a first shaped pulse;
    a second shaping circuit (32,35), configured to shape the measurement pulse according to a second shaping time period (5t 2 ), greater than the first shaping duration, so as to generating a second shaped pulse;
    a comparison unit (44), adapted to compare the first shaped pulse and the second shaped pulse, so as to determine a comparison factor (k);
    a calculating unit (46), configured to determine a correction factor (η) based on the comparison factor (k);
    a correction unit (48), capable of applying the correction factor determined by the calculation unit to the first shaped pulse, so as to form a corrected pulse.
    [16" id="c-fr-0016]
    16. An electronic circuit according to claim 15, in which:
    the first shaping circuit (31,34) has a first delay line, configured to apply a first delay, corresponding to the first shaping time (Stf), to the measurement pulse, the first shaping circuit shaping comprising a subtractor configured to subtract the pulse thus delayed from the measurement pulse;
    the second shaping circuit (32,35) has a second delay line, configured to apply a second delay, corresponding to the second shaping time (5t 2 ), to the measurement pulse, the second circuit a shaping machine comprising a subtractor configured to subtract the pulse thus delayed from the measurement pulse.
    [17" id="c-fr-0017]
    17. Electronic circuit according to claim 16, in which:
    the first shaping circuit is a filter able to generate, from the measurement pulse, a first pulse of predetermined shape parameterized by the first shaping duration (5 ^);
    the second shaping circuit is a filter capable of generating, from the measurement pulse, a second pulse of predetermined shape parameterized by the second shaping duration (5t 2 ).
    [18" id="c-fr-0018]
    18. Electronic circuit according to any one 15 to 17, in which the calculation unit is capable of implementing steps i) to iii) of the method according to any one of claims 1 to 15 from pulses , called pulses of interest, formed by the detector (10) during a time range of interest (Δί)
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3432035B1|2021-08-25|Pulse processing method and electronic circuitry for a pulse generated by an ionizing radiation detector
    EP2484011B1|2013-08-07|Device for the processing of a signal generated by a radiation detector
    EP2507652B1|2013-09-18|Method for calibrating an x-ray detector
    EP2541280B1|2019-01-16|Device for detecting ionising radiation with semiconductor detector with improved spectrometric response
    EP1565763B1|2007-08-22|Enhanced processing circuit for spectrometry system and spectrometry system using same
    EP2510338B1|2014-10-29|Method and device for recognizing a material using the transmission function thereof
    FR2925170A1|2009-06-19|DEVICE FOR REALIZING THE PROCESSING AND SCANNING OF AN ENERGY SPECTRUM OF ELECTROMAGNETIC RADIATION
    EP1004040A1|2000-05-31|SPECTROMETRY MEASURING DEVICE IN THE FIELD OF PHOTON $g| DETECTION
    EP0907086B1|2003-07-23|Device for measuring the rise-time of signals containing noise, arriving from gamma- or x-ray detectors
    EP1598680A2|2005-11-23|System for acquisition of spectra with an integrator having an adaptive threshold
    EP2329511A1|2011-06-08|System for monitoring a photomultiplier gain drift and related method
    EP3532874B1|2020-11-25|Spectrometry device
    EP3237932B1|2019-08-21|Method for calibrating an ionising radiation detector and associated device
    EP3472648B1|2021-08-04|Method and device for characterising an x-ray beam quality
    EP3842840A1|2021-06-30|Method for determining a bias affecting pixels of a pixellated ionising radiation detector
    EP3492950B1|2021-07-21|Method for calibration of a gamma spectrometer, associated calibration system and gamma spectrometry assembly
    EP2605409A1|2013-06-19|Asynchronous digitisation of transient signals from radiation detectors
    EP3111250A1|2017-01-04|Tool for detecting photonic radiation particularly suitable for high-flux radiation
    WO2013014132A1|2013-01-31|Processing device and method for the spectrometric measurement of a photon flux
    同族专利:
    公开号 | 公开日
    FR3069066B1|2019-08-16|
    EP3432035B1|2021-08-25|
    US20190033469A1|2019-01-31|
    US10641909B2|2020-05-05|
    EP3432035A1|2019-01-23|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    CA2732522A1|2008-08-01|2010-02-04|Pulsetor, Llc|Pileup rejection in an energy-dispersive radiation spectrometry system|
    US20170086761A1|2015-09-29|2017-03-30|General Electric Company|Apparatus and method for pile-up correction in photon-counting detector|
    KR100841433B1|2006-06-19|2008-06-25|삼성전자주식회사|Polar transmitter apply to bpsk modulation method using distributed active transformer|
    CN102498412B|2009-09-14|2015-04-08|三菱电机株式会社|Ultrasonic detector|
    FR2951036A1|2009-10-01|2011-04-08|Commissariat Energie Atomique|DEVICE FOR PROCESSING A SIGNAL DELIVERED BY A RADIATION DETECTOR|
    FR2953298B1|2009-11-30|2014-10-31|Commissariat Energie Atomique|METHOD OF CORRECTING THE STACK PHENOMENON APPLIED TO ACQUIRED X-RAY SPECTRA USING A SPECTROMETRIC SENSOR|
    FR2977328B1|2011-07-01|2013-08-02|Commissariat Energie Atomique|IONIZING RADIATION DETECTION DEVICE HAVING AN IMPROVED SPECTROMETRIC RESPONSE RESPONSE SEMICONDUCTOR SENSOR|
    FR3069066B1|2017-07-17|2019-08-16|Commissariat A L'energie Atomique Et Aux Energies Alternatives|METHOD FOR TREATING AN IMPULSE GENERATED BY A IONIZING RADIATION DETECTOR|
    US10324202B1|2018-01-02|2019-06-18|General Electric Company|Systems and methods for collecting radiation detection|FR3069066B1|2017-07-17|2019-08-16|Commissariat A L'energie Atomique Et Aux Energies Alternatives|METHOD FOR TREATING AN IMPULSE GENERATED BY A IONIZING RADIATION DETECTOR|
    JP6808693B2|2018-08-16|2021-01-06|日本電子株式会社|X-ray analyzer and counting rate correction method|
    DE102020208000A1|2020-06-29|2021-12-30|Siemens Healthcare Gmbh|Photon counting X-ray detector, medical imaging device and method for generating an X-ray image data set|
    法律状态:
    2019-01-18| PLSC| Publication of the preliminary search report|Effective date: 20190118 |
    2019-07-31| PLFP| Fee payment|Year of fee payment: 3 |
    2020-07-31| PLFP| Fee payment|Year of fee payment: 4 |
    2021-07-29| PLFP| Fee payment|Year of fee payment: 5 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1756775|2017-07-17|
    FR1756775A|FR3069066B1|2017-07-17|2017-07-17|METHOD FOR TREATING AN IMPULSE GENERATED BY A IONIZING RADIATION DETECTOR|FR1756775A| FR3069066B1|2017-07-17|2017-07-17|METHOD FOR TREATING AN IMPULSE GENERATED BY A IONIZING RADIATION DETECTOR|
    EP18183519.0A| EP3432035B1|2017-07-17|2018-07-13|Pulse processing method and electronic circuitry for a pulse generated by an ionizing radiation detector|
    US16/037,456| US10641909B2|2017-07-17|2018-07-17|Method for processing a pulse generated by a detector of ionizing radiation|
    [返回顶部]