Calculating time to go and size of an object based on scale correlation between images from an electro optical sensor
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06K-009/00
G08G-005/00
G08G-005/04
출원번호
US-0257416
(2009-03-18)
등록번호
US-8774457
(2014-07-08)
국제출원번호
PCT/SE2009/050279
(2009-03-18)
§371/§102 date
20120104
(20120104)
국제공개번호
WO2010/107347
(2010-09-23)
발명자
/ 주소
Jonsson, Jimmy
출원인 / 주소
SAAB AB
대리인 / 주소
Venable LLP
인용정보
피인용 횟수 :
0인용 특허 :
8
초록▼
A method and a system for calculating a time to go value between a vehicle and an intruding object. A first image of the intruding object at a first point of time retrieved. A second image of the intruding object at a second point of time is retrieved. The first image and the second image are filter
A method and a system for calculating a time to go value between a vehicle and an intruding object. A first image of the intruding object at a first point of time retrieved. A second image of the intruding object at a second point of time is retrieved. The first image and the second image are filtered so that the first image and the second image become independent of absolute signal energy and so that edges become enhanced. An X fractional pixel position and a Y fractional pixel position are set to zero. The X fractional pixel position denotes a horizontal displacement at sub pixel level and the Y fractional pixel position denotes a vertical displacement at sub pixel level. A scale factor is selected. The second image is scaled with the scale factor and resampled to the X fractional pixel position and the Y fractional pixel position, which results in a resampled scaled image. Correlation values, are calculated between the first image and the resampled scaled image for different horizontal displacements at pixel level and different vertical displacements at pixel level for the resampled scaled image. A maximum correlation value at a subpixel level is found based on the correlation values. The X fractional pixel position and the Y fractional pixel position are also updated. j is set to j=j+1 and scaling of the second image, calculation of correlation values, finding the maximum correlation value and setting of j to j=j+1 are repeated a predetermined number of times. i is set to i=i+1 and selecting the scale factor, scaling of the second image, calculation of correlation values, finding the maximum correlation value, setting of j to j=j+1, and setting of i to i=i+1 are repeated a predetermined number of times. A largest maximum correlation value is found among the maximum correlation values and the scale factor associated with the largest maximum correlation value. The time to go is calculated based on the scale factor.
대표청구항▼
1. A method for calculating a Time To Go, TTG, value between a vehicle and an intruding object, said method comprising: retrieving a first image of said intruding object at a first point of time, T0, and a second image of said intruding object at a second point of time, T1;filtering said first image
1. A method for calculating a Time To Go, TTG, value between a vehicle and an intruding object, said method comprising: retrieving a first image of said intruding object at a first point of time, T0, and a second image of said intruding object at a second point of time, T1;filtering said first image and said second image so that said first image and said second image become independent of absolute signal energy and so that edges become enhanced;setting an X fractional pixel position, XFRAC, to zero and an Y fractional pixel position, YFRAC, to zero, where XFRAC denotes a horizontal displacement at sub pixel level and YFRAC a vertical displacement at sub pixel level; selecting a scale factor, Si;scaling said second image with said scale factor, Si, and resampling said scaled image to position XFRAC and YFRAC; resulting in a resampled scaled image, RSiI;calculating correlation values, CXPIX, YPIX, i, between said first image and said resampled scaled image, RSiI, for different horizontal displacements at pixel level, XPIX, and different vertical displacements at pixel level, YPIX, for said resampled scaled image RSiI;finding a maximum correlation value at subpixel level, Ci, based on said correlation values, CXPIX, YPIX i, and updating XFRAC and YFRAC;setting j=j+1 and scaling of the second image, calculation of correlation values, finding the maximum correlation value and setting of j to j=j+1 are repeated a first predetermined number of times;setting i=i+1 and selecting the scale factor, scaling of the second image, calculation of correlation values, finding the maximum correlation value, setting of j to j=j+1, and setting of i to i=i+1 are repeated a second predetermined number of times;finding a largest maximum correlation value, CMAX, among said maximum correlation values, Ci, and the scale factor Si, MAX associated with the largest maximum correlation value CMAX; andcalculating the Time To Go, TTG, based on said scale factor Si, MAX. 2. The method according to claim 1, wherein the Time To Go, TTG is calculated as inversely proportional to the scale factor Si , MAX. 3. The method according to claim 1, wherein the Time To Go, TTG is calculated using the formula: TTG=Si, MAX*(T1−T0)/(1−Si, MAX). 4. The method according to claim 1, further comprising, prior to retrieving the first image and the second image, detecting said object in a first sequence of images from a camera creating a second sequence of images in which images are associated with information about detected objects, and wherein said retrieving the first image and the second image comprises retrieving said first and said second images from the second sequence of images. 5. The method according to claim 1, further comprising, prior to retrieving the first image and the second image, detecting said object in a first sequence of images from a camera creating a second sequence of images in which each images are associated with information about detected objects, and after said step of detecting, tracking said object in said second sequence of images, creating a third sequence of images in which said object is centralized in the images, and wherein said retrieving the first image and the second image comprises retrieving said first and said second images from the third sequence of images. 6. The method according to claim 4, wherein said method further comprises, prior to said calculating the Time To Go, TTG, estimating at least one initial size, σXin, σYin, of said object in an image n in said second sequence of images between T0 and T1. 7. The method according to claim 6, wherein said method further comprises estimating at least one initial amplitude, Ain, after said estimating an initial size, of a Gauss function by calculating a difference between a mean of a background and a mean of said object in said image n in the third sequence of images between T0 and T1. 8. The method according to claim 7, wherein said method further comprises calculating at least one size, σXn, σYn, after said step of estimating an initial amplitude, by determining a Gaussian function, Gn, so that an error between said Gaussian function, Gn, and said object in said image n in the third sequence of images between T0 and T1 is minimized, where said initial amplitude Ain and said initial size, σXin, σYin are used as start parameters for said Gaussian function. 9. The method according to claim 8, wherein said method of calculating a size further comprises filtering several sizes, σXn, σYN, from several images and thereby achieving a filtered size, σXF, σYF. 10. The method according to claim 8, wherein said method further comprises updating said size σXn+1, σYn+1 for consecutive images n+1 in said third sequence based on said size σXn, σYn in said image n by using formulas: σXn+1=σXn(TTGn+1+1/f)/TTGn+1 σYn+1=σYn(TTGn+1+1/f)/TTGn+1 where f is an image frequency between consecutive images. 11. The method according to claim 10, wherein in said updating the size said filtered size, σXF, σYF, are used as start value. 12. A computer program product for use in a vehicle for calculating a Time To Go, TTG, between said vehicle and an intruding object, the computer program product comprising a non-transitory computer readable medium, having thereon: computer readable code which, when run in a processor of the vehicle causes the processing means to perform;retrieving a first image of said intruding object at a first point of time, T0, and a second image of said intruding object at a second point of time, T1;filtering said first image and said second image so that said first image and said second image become independent of absolute signal energy and so that edges become enhanced;setting an X fractional pixel position, XFRAC, to zero and an Y fractional pixel position, YFRAC, to zero, where XFRAC denotes a horizontal displacement at sub pixel level and YFRAC a vertical displacement at sub pixel level; selecting a scale factor, Si;scaling said second image with said scale factor, Si, and resampling said scaled image to position XFRAC and YFRAC; resulting in a resampled scaled image, RSiI;calculating correlation values, CXPIX, YPIX, i, between said first image and said resampled scaled image, RSiI, for different horizontal displacements at pixel level, XPIX, and different vertical displacements at pixel level, YPIX, for said resampled scaled image RS1I;finding a maximum correlation value at subpixel level, Ci, based on said correlation values, CXPIX, YPIX i, and updating XFRAC and YFRAC;setting j=j+1 and scaling of the second image, calculation of correlation values, finding the maximum correlation value and setting of j to j=j+1 are repeated a first predetermined number of times;setting i=i+1 and selecting the scale factor, scaling of the second image, calculation of correlation values, finding the maximum correlation value, setting of j to j=j+1, and setting of i to i=i+1 are repeated a second predetermined number of times;finding a largest maximum correlation value, CMAX, among said maximum correlation values, Ci, and the scale factor Si, MAX associated with the largest maximum correlation value CMAX; andcalculating the Time To Go, TTG, based on said scale factor Si, MAX. 13. The computer program product according to claim 12, wherein the Time To Go, TTG is calculated as inversely proportional to the scale factor Si, MAX. 14. The computer program product according to claim 12, wherein the Time To Go, TTG is calculated using the formula: TTG=Si, MAX*(T1−T0)/(1−Si, MAX). 15. The computer program product according to claim 12, wherein said computer readable code when run in said processor further causes the processor to perform; prior to said retrieving the first image and the second image, detecting said object in a first sequence of images from a camera creating a second sequence of images in which images are associated with information about detected objects, wherein said retrieving further comprises retrieving said first and said second images from the second sequence of images. 16. The computer program product according to claim 12, wherein said computer readable code when run in said processor further causes the processor to perform detecting said object in a first sequence of images from a camera creating a second sequence of images in which images are associated with information about detected objects; and after said detecting, andtracking said object in said second sequence of images, creating a third sequence of images in which said object is centralized in the images,wherein said retrieving further comprises retrieving said first and said second images from the third sequence of images. 17. The computer program product according to claim 15, wherein said computer readable code when run in said processor further causes the processor to perform, prior to said calculating the Time To Go, TTG; estimating at least one initial size, σXin, σYin, of said object in an image n in said second sequence of images between T0 and T1. 18. The computer program product according to claim 16, wherein said computer readable code when run in said processor further causes the processor to perform, after said estimating an initial size; estimating at least one initial amplitude, Ain, of a Gauss function by calculating a difference between a mean of a background and a mean of said object in said image n in the third sequence of images between T0 and T1. 19. The computer program product according to claim 18, wherein said computer readable code when run in said processor further causes the processor to perform, after said estimating an initial amplitude; calculating at least one size, σXn, σYn, by determining a Gaussian function, Gn, so that an error between said Gaussian function, Gn, and said object in said image n in the third sequence of images between T0 and T1 is minimized, wherein said initial amplitude Ain and said initial size, σXin, σYin are used as start parameters for said Gaussian function. 20. The computer program product according to claim 19, wherein said computer readable code when run in said processor further causes the processor to perform in said calculating at least one size filtering of several sizes, σXn, σYn, from several images and thereby achieving a filtered size, σXF, σYF. 21. The computer program product according to claim 18, wherein said computer readable code when run in said processor further causes the processor to perform; updating said size σXn+1, σYn+1 for consecutive images n+1 in said third sequence based on said size σXn, σYn in said image n by using formulas: σXn+1=σXn(TTGn+1+1/f)/TTGn+1 σYn+1=σYn(TTGn+1+1/f)/TTGn+1 where f is the image frequency. 22. The computer program product according to claim 21, wherein said computer readable code when run in said processor further causes the processor to use said filtered size, σXF, σYF as start values in said updating the size. 23. A system for calculating a Time To Go, TTG, value between a vehicle and an intruding object, said system comprises comprising: a memory module comprising a computer program product according to claim 12; anda processor configured for running said computer program product.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (8)
Coleman, Norman P.; Lin, Ching-Fang, 4D GIS based virtual reality for moving target prediction.
Pfeiffer Carl G. ; Tsai Cheng-Chih ; Gumas D. Spyro ; Calingaert Christopher ; Nguyen Danny D., Background adaptive target detection and tracking with multiple observation and processing stages.
Kurata, Tohru, Image processing device, image processing method, image pickup device, and image pickup method for superimposing a plurality of pictures on each other.
Numazaki Shunichi,JPX ; Doi Miwako,JPX ; Morishita Akira,JPX ; Umeki Naoko,JPX ; Miura Hiroki,JPX, Method and apparatus for generating information input using reflected light image of target object.
Banh Nam D. (Canoga Park CA) Lo Thomas K. (Temple City CA) Holthaus Kelly D. (Chatsworth CA) Sacks Jack M. (Thousand Oaks CA), Moving target detection method using two-frame subtraction and a two quadrant multiplier.
Newman, Arthur M.; Kaganovich, Vitaliy M.; Sun, Nicholas; Douglas, Dave S., Three-frame difference moving target acquisition system and method for target track identification.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.