专利摘要:
Imaging device for photofinish type system, comprising a first camera aligned with a finishing line, and characterized in that it comprises means for displaying images in three dimensions of stereoscopic type.
公开号:AU2013224324A1
申请号:U2013224324
申请日:2013-02-11
公开日:2014-09-04
发明作者:Pascal Richard
申请人:Swiss Timing Ltd;
IPC主号:G03B15-00
专利说明:
Case 3349 BG Device for 3D display of photo finish image Field of the invention The present invention concerns the display of timed data during a 5 sports competition, and more specifically a photo finish type at a finish line. Background of the invention In timed sports races or competitions, and in particular athletics races, the timing device is generally started in a synchronised manner with 10 the emission of an acoustic signal, and stopped for each competitor when each competitor respectively crosses the finish line. It may happen however that the resolution of the timing system, for example typically a hundredth of a second, is insufficient to distinguish between two competitors whose measured time is identical. To achieve this, auxiliary devices are required to 15 assist timing and there are manually operated systems for this purpose which are based on the visual recognition of images taken in sequence by a high definition camera very precisely centred on the finish line. An operator analyses, in an asynchronous manner after the competitors have finished the race, the sequence of images taken at given times and consequently 20 corresponding to a measured time; this analysis therefore makes it possible to determine a more accurate line crossing time for each competitor, for example to a thousandth of a second, and to rank the competitors in a reliable manner. This type of auxiliary timing display is usually called a "photo finish" device. 25 This type of auxiliary display system, known for a relatively long time, is disclosed for example in US Patent No. 3502009 relating to a dog race. The photo finish device enables time values, obtained using a first lens - 2 centred on a timing device, to be superimposed on images of the competitors using a second lens centred on the finish line. The developed film thus makes it possible to determine the first dog to cross the line and its lap or finish time. 5 To overcome the problem of competitors masking each other, which may prevent the operator performing a pertinent analysis of the sequence of images around the disputed measured time, photo finish systems very often use several cameras. However, because of the precise alignment required on the finish line, none of the cameras used can be shifted laterally 10 to resolve this masking problem; this is why the two cameras employed are respectively located on either side of the finish line. There are also known stereoscopic optical systems for the display of three dimensional images, which provide a more user-friendly and intuitive rendering for the person viewing the images. These stereoscopic systems 15 are based on the construction of an image in relief in the same manner as the brain, i.e. from the analysis of two plane images taken along distinct axes of view. These stereoscopic systems thus typically use dual lens cameras, the space between the lenses corresponding, for example, to the mean interocular space, namely around 75 to 80 millimetres. However, 20 these optical systems are unfortunately unsuitable for a photo finish system because of the lateral offset that is required between the photograph shooting axes, which cannot be simultaneously aligned on the finish line and therefore there is no possibility of these systems overcoming the shortcomings of the timing system in a reliable and accurate manner. 25 There is consequently a need for auxiliary timing systems which are free of these known limitations. Brief summary of the invention It is an object of the present invention to provide a device and a 30 method for the display of three dimensional photo finish images.
- 3 It is another object of the present invention to provide a device and a method of this type which are simple to implement and inexpensive. These objects are achieved, in particular, as a result of an imaging device for a photo finish system, including a first camera aligned on a finish 5 line, and characterized in that the device includes a second camera also aligned on the finish line and offset by a first height in relation to the first camera for the display of three dimensional stereoscopic images. These objects are also achieved as a result of a first method for forming a three dimensional photo finish image, characterized in that it 10 contains the steps of defining a shot duration T, of time synchronising a first and a second camera aligned on the finish line and offset by a height h, on a reference time To, of forming a first image by the first camera for reference time To and of forming a second image by the second camera for the same reference time To, and then aggregating these two images to 15 form a stereoscopic image. Specific embodiments of the invention are defined in the dependent claims. One advantage of the present invention is in producing a photo finish image in relief in a simple manner, without requiring complex image 20 processing tools, and which can easily be displayed to spectators present in a sports arena or in front of their TV screen. Another advantage of the solution is to provide a modular architecture for the implementation of a stereoscopic display system and method, which requires few modifications to ordinary existing infrastructures 25 for photo finish systems. Brief description of the drawings Advantageous example implementations of the invention are given in the description and illustrated in the annexed Figures, in which: - Figures 1A and B respectively show a front view, in the axis of a finish line, and a top view of a plurality of cameras used in a preferred embodiment of the invention; - Figure 2 shows a diagram of the cameras used and an image 5 management system for the production of stereoscopic images according to a preferred embodiment of the invention; - Figure 3 shows a diagram of the various processing steps for forming and displaying images according to a preferred embodiment of the invention. 10 Example embodiment(s) of the invention Figure 1A shows a front view, in the axis of a finish line, of a plurality of cameras used in a preferred embodiment of the present invention. This Figure shows the various lanes L1, L2, L3, L4, L5, L6, L7, L8, which are the 15 normal lanes of an athletics track, and four cameras C1, C2, C3, C4 arranged at a height on each side of the athletics track. According to this preferred embodiment, the stereoscopic images are formed from two images taken at the same reference moment by both cameras which are offset vertically from each other. Thus a first camera C1 and a second 20 camera C2, both aligned on the finish line, are offset from each other by a first height h and similarly a third camera C3 and a fourth camera C4, both aligned on the finish line, are offset from each other by a second height h'. The advantage of arranging two pairs of cameras located on either side of the finish line, i.e. in forming pairs C1-C2 and C3-C4, consists in the ability 25 to choose the most advantageous camera pair according to the position of the various competitors on the finish line, to avoid any masking. A first pair of cameras, for example the pair C1-C2 could be attributed the status of main pair, i.e. the default pair of cameras used by the imaging device of the invention, while the second pair of cameras C3-C4 could be used as an 30 auxiliary pair only if competitors are masked in images obtained from the - 5 first pair of cameras C1 -C2. Figure 1 B illustrating the same eight lanes Li L8 and finish line 0, shows the alignment of the four cameras C1,C2,C3,C4 on finish line 0 and their grouping in pairs C1-C2 and C3-C4 on either side of finish line 0. The cameras of each pair are shown with a slight axial offset 5 so that they can be distinguished more easily from each other. The height offset between the cameras of each pair is chosen, for practical reasons, to be considerably greater than the interocular space to facilitate the positioning of the cameras. In practice, a first space h and a second space h' preferably comprised between 40 and 50 centimetres 10 could be selected, to leave sufficient space to mount the second camera of the pair after the first has been placed in position, and, at the same time, to enable the multiplicative factor of the distance between the cameras of each pair in relation to the mean interocular space to be corrected by zoom effect, i.e. by modifying the focal distance of the camera lens. The first 15 space h and the second space h' do not, however, necessarily need to be equal, and could be adjusted according to the constraints of the place where the competition takes place, to avoid requiring any adjustment to the existing infrastructure and thus to minimise the space constraints of the proposed imaging system. 20 Figure 2 shows a diagram of the imaging device for a photo finish system according to a preferred embodiment of the invention, showing in detail the physical architecture of the two cameras C1 and C2, illustrated in Figure 1 B above, and the interaction thereof with the timing system 6 and the stereoscopic display system 3. 25 Each of the two cameras C1 and C2 includes a lens (respectively referenced 10 and 20 for each camera) at the end of a lens assembly for forming the narrow strip image (respectively referenced 1 and 2 for each camera, and also commonly called "frames") of height H and width / via line-scan-charge coupled devices (LS-CCD), or CMOS respectively 30 referenced 11 and 21 for each camera C1 and C2, whose size may be up to several thousand pixels. The lens assembly is preferably adjusted on the - 6 finish line to display an image of a predefined height H, which is achieved using a number N of pixels of the matrix seen in Figures 3A and 3B below. The LS-CCD/CMOS sensor may preferably consist of a matrix with a single column of pixels, so that the corresponding width / of frames 1 and 2 is 5 determined entirely by height H of the image and the number N of pixels of the column of said sensor. According to a variant, a CCD/CMOS matrix sensor having several columns may be used, only one predefined column of which is used to form the images at the finish line. An image sampler/controller (respectively referenced 15-25 on each 10 camera C1, C2) senses the lighting level at each of LS-CMOS sensors 11 21 at regular time intervals, then amplifies the output signals using gain controlled amplifiers (respectively referenced 12 and 22) on each of cameras C1-C2) and digitally converts them using analogue-digital converters (A/D, respectively referenced 13-23 for each of cameras C1 and 15 C2) and saves them in a buffer (respectively referenced 14-24 for each of cameras C1 and C2). Thus a series of frames 1 and 2 respectively taken by each of cameras C1 and C2 are saved in the form of pixel columns 101 and 102 respectively (seen in Figure 3A below), at a sampling frequency preferably higher than or equal to 1000, so that images corresponding to 20 references times separated by a maximum of a thousandth of a second can be formed by aggregating the different pixel columns 101, 102. The sampling frequency thus defines timing accuracy. The buffers preferably contain several hundred megabytes of RAM, for example SDRAM (synchronous dynamic random access memory) for containing several 25 hours of images. To guarantee synchronisation between each of cameras C1 and C2 on a relative reference time, i.e. in relation to the start of a race, and not only an absolute time, each camera will be connected to timing system 6. Each of the transmitted pixel columns corresponding to a reference time 30 can be saved with a flag corresponding to the reference time and thereby allow for easier subsequent data processing.
The first camera C1 and second camera C2 are each connected to an image management system 4, such as for example a computer server, including a processor 41 and a memory unit 42 for processing data recovered at the output of buffers 14-24 of cameras C1 and C2 and to form 5 a stereoscopic image 30. Image management system 4 is also connected to a monitoring screen 5 to allow an operator to view the images before transmitting them, via a transmission line 31, for example a broadcast line, to television receivers (for example via a digital hertz network conforming to DVB standard) or to a group of computers connected to the Internet via an 10 IP-multicast network. Thus the stereoscopic display system 3 illustrated in Figure 2, which includes image management system 4 and the connection to transmission line 31, can firstly form a stereoscopic image 30 from frames received from cameras C1 and C2, and also transmit this image for display on remote 15 receivers (not shown in this Figure). It will be understood that a similar physical architecture to that described in Figure 2 will preferably be used for cameras C3 and C4 disposed on the opposite side of the finish line, which will also be connected to computer image management system 4 and to timing system 20 6. These connections may be based on a Wi-Fi type wireless technology depending on the distance between the devices to be connected and the required data transmission flow; for reasons of compactness it will be sought to avoid using cables when unnecessary. Figure 3 shows a preferred variant of the method for generating a 25 stereoscopic photo finish image 3. The method for forming a photo finish image in three dimensions contains a preliminary step (A) of defining a shot duration (T), which determines the size of the image required to be displayed, or more specifically the overall width L thereof. Indeed, if each image strip or frame has a unitary width /, the sampling frequency, for 30 example every thousandth of a second, and the shot duration T, which is chosen to be a multiple of a thousandth, such as for example one second, - 8 determines a number of pixels M which defines the overall width L of the corresponding image. In these Figures, the time axis t extends from right to left from a selected reference moment To which is the moment at which the frames the furthest to the right are taken, all the other frames having been 5 taken subsequent to this reference moment. It will be understood however that if cameras C3 and C4 were used, with the direction of the race thus reversed, the time axis t would extend from left to right, i.e. still in the opposite direction to the direction of the race. As seen in Figure 3, the first and second images taken by first and 10 second cameras C1 and C2 are respectively referenced 11 and 12; they are digitised in the form of a pixel matrix N*M, corresponding to a height H and a width L. The pixel matrix is obtained by aggregating the various pixel columns 101 and 102 respectively corresponding to frames 1 and 2. In order for each image 11 and 12 of each camera C1 and C2 to 15 correspond to the same moment To, a time synchronisation step (BO) must be carried out on first camera C1 and second camera C2; both cameras are aligned on finish line 0 before being able to form stereoscopic image 30, since, according to this preferred embodiment each of the images forming the stereoscopic image must correspond to what each eye sees 20 simultaneously. Once this synchronisation has been performed as a result of the connection of each camera to timing system 6, there is formed a first image 11, obtained from said first camera C1 for said reference time To and a second image L2, obtained from second camera C2 for reference time TO in steps B1 and B2, produced alternately or simultaneously by image 25 management system 4. Before aggregating the first and second images (11, 12) to form a stereoscopic image 30, an additional image rectifying step C may be carried out to make the displayed images easier to read. Indeed, the axis of the two cameras C1, C2 - and also that of cameras C3, C4 on the opposite side of 30 the finish line - is vertical in the plane of the finish line, and it determines the line of orientation of the eyes perceiving the stereoscopic image. In other - 9 words, the stereoscopic image is formed as if first camera C1 were the left or right eye, and second camera C2 the other eye, which would force a television viewer to turn his head 90 degrees to the right or left to see stereoscopic image 30 properly, formed by images 11 and 12, without any 5 other preliminary image processing. Thus, according to a preferred variant of the invention, an additional image rectifying step C will be carried out, consisting, for example, of reverse stacking C' and matrix transposition C" sub-steps when the rectifying operation is carried out by computer image processing system 4. The sorting sub-step C' allows central axial symmetry 10 to be achieved by saving all the pixel columns corresponding to the frames of images 11 and 12 in LIFO mode (last in first out) in a new data stack, whereas the matrix transposition operation consists in reversing the lines and columns of the matrix. It can be easily appreciated that the step sequence of C' followed by C" allows each of the images to be rotated 90 15 degrees in the anti-clockwise direction, while the step sequence of C" followed by C' allows a 90 degree rotation in the clockwise direction. The choice of one sequence or the other thus virtually assigns one camera to one eye and determines the axis of view: step C' followed by step C" will associate first camera C1 with the right eye and will rectify images 11 and 12 20 to produce a stereoscopic image 30 with a front view of the competitors and the reverse sequence, i.e. step C" followed by step C' will associate first camera C' with the left eye with a back view of the competitors. This is why the images obtained after rectifying step C are reference 11/12 and 12/11 in Figure 2, depending on the selected sequence. 25 Once this rectifying step has been performed, stereoscopic image 30 can be formed by computer image management system 4 by aggregating two rectified images in aggregation step D, then, in a subsequent step E, stereoscopic display system 3 can emit stereoscopic image 30 on a transmission network 31, for example a broadcast network, for remote 30 display on remote receivers (e.g. TV) after verification by an operator on a monitoring screen. It will be clear that, to determine reference time To, the photo finish system operator could perform preliminary display iteration on - 10 monitoring screen 5; this preliminary step could also establish any masking on first camera C1 and the need to switch to the pair of third and fourth cameras C3, C4 on the opposite side of the finish line. According to the preferred embodiment illustrated, digital processing 5 means are therefore used to rectify the image before forming a stereoscopic image via computer image management system 4; according to an alternative embodiment it is possible to envisage using periscope type optical means with a 45 degree return, which would remove the need for the data processing step. To avoid any modification of the existing structure of 10 cameras used at the finish line, periscope glasses, mounted in a modular manner on the lens, could be used. Although the solutions for forming stereoscopic images have been described with reference to athletics competitions, it will be understood that the device and method described above may be employed for other types 15 of sport, such as, for example cycling, speed skating or rowing, without departing from the scope of the invention. The above device and method also have the advantage of only requiring very few modifications of the physical architecture of existing photo finish systems, with the mere possible addition of extra cameras for 20 capturing images from other angles. Thus, the entire intelligence of the system and method proposed is transferred in the back-end of the infrastructure, i.e. in a part invisible to spectators, with additional modules for post-processing the conventional photo finish images obtained by the camera or cameras at the finish line, and specific stereoscopic display 25 means for forming three dimensional images.
权利要求:
Claims (10)
[1] 1. Imaging device for a photo finish type system, including a first camera (C1) aligned on a finish line (0), and characterized in that the device includes: 5 a second camera (C2) also aligned on said finish line (0) and offset by a first height (h) from said first camera (Cl) for the display of three dimensional stereoscopic images.
[2] 2. Imaging device for a photo finish system according to the preceding claim, characterized in that the height offset (h) between said first 10 and second cameras (C1, C2) is greater than the interocular space.
[3] 3. Imaging device for a photo finish system according to any of the preceding claims, characterized in that the device includes a third camera (C3) also aligned on said finish line (0), arranged on the opposite side of said finish line (0) with respect to said first camera (Cl), and a fourth 15 camera (C4) also aligned on said finish line (0) and arranged on the same side of the finish line as said third camera (C3), said fourth camera (C4) being offset by a second height (h') from said third camera (C3).
[4] 4. Imaging device for a photo finish system according to claim 1, characterized in that the device further includes image rectifying means, 20 said image rectifying means being optical or digital processing type means.
[5] 5. Imaging device for a photo finish system according to any of the preceding claims, characterized in that each of the first and second cameras (C1, C2) uses line scan (LS-CCD/CMOS) sensors (11, 21)
[6] 6. Imaging device for photo finish system according to any of the 25 preceding claims, characterized in that the device includes a computer image management system (4) connected, on the one hand, to each of said first and second cameras (C1, C2) to form a stereoscopic image (30) and, - 12 on the other hand, to a monitoring screen (5) and to a transmission network (31).
[7] 7. Method for forming a three dimensional photo finish image, characterized in that the method contains the steps of: 5 - (A) defining a shot duration (T); - (BC) time synchronising a first camera (Cl), aligned on a finish line (0), and a second camera (C2), aligned on said finish line (0), wherein said first and second cameras (C1, C2) is offset by a height (h) on said finish line (0), on a reference time To; 10 - (B1) forming a first image (11) obtained from said first camera (Cl) for said reference time To; - (B2) forming a second image (12) obtained from said second camera (C2) for said reference time To; - (D) aggregating said first and second images (11, 12) to form a 15 stereoscopic image (30).
[8] 8. Method according to claim 7, characterized in that the method contains an additional image rectifying step (C).
[9] 9. Method according to claim 8, said image rectifying step (C) including a first reverse stacking sub-step (C') and a second matrix 20 transposition sub-step (C").
[10] 10. Method according to any of claims 7 to 9, characterized in that the method includes an additional step: - (E) transmitting said stereoscopic image (30) on a transmission network (31) for remote display on remote receivers. 25
类似技术:
公开号 | 公开日 | 专利标题
CN1934874B|2010-07-21|Three dimensional acquisition and visualization system for personal electronic devices
US20130286170A1|2013-10-31|Method and apparatus for providing mono-vision in multi-view system
US8897502B2|2014-11-25|Calibration for stereoscopic capture system
CN101883215A|2010-11-10|Imaging device
CN101636747A|2010-01-27|Two dimensional/three dimensional digital information obtains and display device
TW201205308A|2012-02-01|Combining data from multiple image sensors
AU2013224324B2|2016-01-28|Device for 3D display of a photofinish image
WO2016045425A1|2016-03-31|Two-viewpoint stereoscopic image synthesizing method and system
CN104994370A|2015-10-21|3D | video image acquisition system
JP4424299B2|2010-03-03|Image processing method, image processing apparatus, and image display apparatus using the same
CN105472308A|2016-04-06|Multi-view naked eye 3D video conference system
US20150116459A1|2015-04-30|Sensing device and signal processing method thereof
WO2001020386A2|2001-03-22|An autostereoscopic display and method of displaying three-dimensional images, especially color images
JP5088973B2|2012-12-05|Stereo imaging device and imaging method thereof
CN206301143U|2017-07-04|A kind of many mesh 3D panorama shooting devices
US20150035988A1|2015-02-05|Video imaging system including a plurality of cameras and a plurality of beamsplitters
CN214756700U|2021-11-16|3D display device
US20220021852A1|2022-01-20|Color fringing processing independent of tone mapping
CN112929639A|2021-06-08|Human eye tracking device and method, 3D display equipment and method and terminal
WO2015062141A1|2015-05-07|Pixel unit of 3d cmos image sensor
US20050057438A1|2005-03-17|Apparatus and method for producing three-dimensional motion picture images
CN102769763B|2016-12-14|3-dimensional image camera and corresponding control methods thereof
TWI469043B|2015-01-11|Interactive image playing system and method
CH706165A2|2013-08-30|Imaging device for photo finish system used to display time data during e.g. athletics, has camera aligned on arrival line, display unit to display three-dimensional images, and another camera shifted from height relative to former camera
WO2014065700A1|2014-05-01|System for producing animated images, for example video images
同族专利:
公开号 | 公开日
US9774842B2|2017-09-26|
JP2015515605A|2015-05-28|
CA2865015A1|2013-08-29|
JP6110883B2|2017-04-05|
US20150015675A1|2015-01-15|
AU2013224324B2|2016-01-28|
WO2013124180A2|2013-08-29|
EP2817675B1|2016-04-20|
HK1204084A1|2015-11-06|
CA2865015C|2017-12-05|
CN104246598A|2014-12-24|
WO2013124180A3|2014-10-23|
EP2817675A2|2014-12-31|
引用文献:
公开号 | 申请日 | 公开日 | 申请人 | 专利标题
US3502009A|1967-12-22|1970-03-24|Fred J Connors|Race timer system|
US4488794A|1982-06-28|1984-12-18|Joanne Dolgow|Photographic apparatus for making simultaneous exposures|
JPH07298307A|1994-04-28|1995-11-10|Canon Inc|Image recording and reproducing device|
JP3183318B2|1994-05-20|2001-07-09|日立電子株式会社|Arrival order and time judgment device|
JP3222333B2|1994-10-17|2001-10-29|セイコーインスツルメンツ株式会社|Arrival order judgment device|
US5974272A|1997-10-29|1999-10-26|Eastman Kodak Company|Parallax corrected image capture system|
US6545705B1|1998-04-10|2003-04-08|Lynx System Developers, Inc.|Camera with object recognition/data output|
US6433817B1|2000-03-02|2002-08-13|Gavin Guerra|Apparatus and method for determining the winner of a race|
JP3729818B2|2003-05-14|2005-12-21|株式会社日立国際電気|Arrival order and time determination device|
JP2006339703A|2005-05-31|2006-12-14|Furoobell:Kk|Image processing apparatus and method, program, and recording medium|
US8442304B2|2008-12-29|2013-05-14|Cognex Corporation|System and method for three-dimensional alignment of objects using machine vision|
US8090251B2|2009-10-13|2012-01-03|James Cameron|Frame linked 2D/3D camera system|
JP2013048334A|2011-08-29|2013-03-07|Sony Corp|Image processor, image processing system and program|
US9390752B1|2011-09-06|2016-07-12|Avid Technology, Inc.|Multi-channel video editing|US9413926B2|2014-04-28|2016-08-09|Lynx System Developers, Inc.|Systems for processing event timing images|
US10375300B2|2014-04-28|2019-08-06|Lynx System Developers, Inc.|Methods for processing event timing data|
US9398196B2|2014-04-28|2016-07-19|Lynx System Developers, Inc.|Methods for processing event timing images|
法律状态:
2016-05-26| FGA| Letters patent sealed or granted (standard patent)|
2020-09-10| MK14| Patent ceased section 143(a) (annual fees not paid) or expired|
优先权:
申请号 | 申请日 | 专利标题
EP12156514||2012-02-22||
EP12156514.7||2012-02-22||
PCT/EP2013/052698|WO2013124180A2|2012-02-22|2013-02-11|Device for 3d display of a photofinish image|
[返回顶部]