IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0661739
(2003-09-12)
|
발명자
/ 주소 |
- Laird, Mark D.
- Small, Steven I.
- Tinnemeier, Michael T.
|
출원인 / 주소 |
|
대리인 / 주소 |
Weingarten, Schurgin, Gagnebin &
|
인용정보 |
피인용 횟수 :
26 인용 특허 :
86 |
초록
▼
A system and method for detecting and filtering non-violation events in a traffic light violation prediction and recording system, including at least one violation prediction image capturing device, such as a video camera, and a violation prediction unit. The prediction unit generates a prediction r
A system and method for detecting and filtering non-violation events in a traffic light violation prediction and recording system, including at least one violation prediction image capturing device, such as a video camera, and a violation prediction unit. The prediction unit generates a prediction reflecting a probability that the vehicle will violate a red light phase of the traffic signal. A non-violation event filter determines whether the vehicle approaching the traffic signal is actually performing a non-violation action. Non-violation events may include a variety of actions performed by the vehicle, and are fully configurable to meet the needs and policies of various specific intersections and jurisdictions. When the non-violation event filter determines that the vehicle is performing a non-violation action, it causes some number of resources that may have been allocated to recording the non-violating vehicle, and/or prevents further resources from being allocated to such recording.
대표청구항
▼
1. A system for detecting a violation of a traffic signal at an intersection comprising the steps of:a virtual violation line interface for receiving from a user data defining a virtual violation line that corresponds to a location at said intersection that if crossed by a vehicle entering said inte
1. A system for detecting a violation of a traffic signal at an intersection comprising the steps of:a virtual violation line interface for receiving from a user data defining a virtual violation line that corresponds to a location at said intersection that if crossed by a vehicle entering said intersection during a red light phase of said traffic signal, is indicative of a violation of said traffic signal by said vehicle; a storage device for storing a representation of said intersection and said virtual violation line; at least one camera for capturing at least one image of a vehicle at said intersection; a processing unit operative to: analyze said at least one image to identify a position of said vehicle with resperct to said virtual violation line, compare said position of said vehicle to said virtual violation line, and generate an indication of a violation in the event said processing unit determines that said position of said vehicle is beyond said virtual violation line and that said vehicle has traveled into said intersection during said red light phase of said traffic signal. 2. The system of claim 1 wherein said at least one camera is further operative to capture a plurality of images of said vehicle approaching said intersection; andsaid processing unit is operative to analyze said plurality of images of said vehicle and to generate a prediction signal in the event it is determined by said processing unit that said vehicle is likely to violate said red light phase of said traffic signal. 3. The system of claim 2 further including an additional signaling device for cross traffic approaching said intersection from a direction other than said vehicle, said additional signaling device responsive to said prediction signal for signaling said cross traffic not to enter said intersection.4. The system of claim 3 wherein said additional signaling device comprises an additional traffic signal and said additional traffic signal is responsive to said prediction signal to delay a green light phase of said additional traffic signal.5. The system of claim 1, wherein said at least one camera comprises at least one video camera.6. The system of claim 1, wherein said at least one camera comprises at least one digital camera.7. The system of claim 1, wherein said processing unit comprises at least one microprocessor.8. The system of claim 2, wherein said processing unit is further responsive to a time remaining in a yellow light phase of said traffic signal in the generation of said prediction signal.9. The system of claim 2, wherein said processing unit is operable to determine a current speed for said vehicle and to utilize said current speed in determining whether to generate said prediction signal.10. The system of claim 2, wherein said processing unit is operable to determine a current acceleration for said vehicle and to utilize said current acceleration in determining whether to generate said prediction signal.11. The system of claim 10, wherein said processing unit is operable to compute a time remaining before said vehicle reaches said location corresponding to said virtual violation line responsive to said determination of said current acceleration.12. The system of claim 11, wherein said processing unit is operable to calculate a deceleration required for said vehicle to stop within said time remaining before said vehicle reaches said location corresponding to said virtual violation line.13. The system of claim 12 wherein said processing unit is operable to determine whether said required deceleration is larger than a specified deceleration limit value, and if so, to generate said prediction signal.14. The system of claim 1, wherein said virtual violation line is stored within said storage device as a portion of said representation of said intersection.15. A method for detecting a violation of a traffic signal comprising the steps of:storing in a storage device a representation of a traffic intersection, said representation of said intersection including a virtual violation line corresponding to a location at said intersection that if crossed by a vehicle entering said intersection during a red light phase of said traffic signal, is indicative of a violation of said traffic signal by said vehicle, said location of said virtual violation line with respect to said intersection being user configurable; capturing at least one image showing said vehicle at said intersection; analyzing said at least one image of said vehicle at said intersection to ascertain a position of said vehicle with respect to said location corresponding to said virtual violation line; and generating an output indicative of a violation of a red light phase of said traffic signal in the event said analyzing step indicates that said vehicle has traveled beyond said location corresponding to said virtual violation line and into said intersection during said red light phase of said traffic signal. 16. The method of claim 15, further comprising the steps of:capturing a plurality of images showing said vehicle approaching said traffic signal; and generating a prediction signal, responsive to said plurality of images, and an indication of a current traffic signal light phase, in response to a determination that said vehicle is likely to violate said red light phase of said traffic signal. 17. The method of claim 16, further including the step of signaling cross traffic approaching said intersection from a direction other than said vehicle not to enter said intersection responsive to said prediction signal.18. The method of claim 17 wherein said step of signaling said cross traffic includes the step of delaying a green phase of a traffic signal for said cross traffic.19. The method of claim 15, wherein said capturing step comprises the step of capturing said at least one image with a video camera.20. The method of claim 15, wherein said capturing step comprises the step of capturing said at least one image with a digital camera.21. The method of claim 16, wherein said step of generating said prediction signal includes the steps of:determining a time remaining for said vehicle in a yellow light phase of said traffic signal; and generating said prediction signal, based in part, upon said time remaining in said current yellow light phase. 22. The method of claim 16 wherein said step of generating said prediction signal includes the steps of:determining from said plurality of images a current speed for said vehicle; and generating said prediction signal, based in part, upon said current speed of said vehicle. 23. The method of claim 16, wherein said step of generating said prediction signal includes the step of generating from said plurality of images a current acceleration for said vehicle and said step of generating said prediction signal is based in part upon said current acceleration of said vehicle.24. The method of claim 16, wherein said step of generating said prediction signal further includes the step of generating a time remaining before said vehicle crosses said location corresponding to said virtual violation line and said step of generating said prediction signal is based, in part, upon said time remaining.25. The method of claim 16, wherein said step of generating said prediction signal includes the step of:calculating a rate of deceleration that is required for said vehicle to stop before said location corresponding to said virtual violation line; and generating said prediction signal in the event said required rate of deceleration is greater than a predetermined deceleration limit value. 26. The method of claim 15, wherein said storing step comprises the step of storing in said storage device a representation of said virtual violation line that corresponds to a location beyond an actual stop line for a lane of traffic in which said vehicle is travelling.27. The method of claim 15, wherein said storing step includes the step of storing said representation of said virtual violation line for locations corresponding to each of a plurality of lanes of traffic approaching said intersection.28. The method of claim 27 further including the step of identifying which one of said plurality of lanes of traffic said vehicle is travelling in from an analysis of said at least one image.29. A collision avoidance system for a first traffic signal having a current light phase equal to one of the set consisting of at least red and green and a second traffic signal having a current light phase equal to one of the set consisting ofat least red and green, comprising: at least one violation image capturing device; a plurality of images showing at least one vehicle approaching said first traffic signal, said images derived from an output of said violation image capturing device. a processing unit responsive to said plurality of images and an indication of said current first traffic signal light phase, for generating at least one violation prediction for said at least one vehicle approaching said first traffic signal, said violation prediction indicating a likelihood that said at least one vehicle approaching said first traffic signal will violate an upcoming red light phase of said first traffic signal; a collision avoidance unit responsive to said violation prediction, for asserting at least one violation predicted signal; and a traffic light controller for said second traffic signal for controlling said second traffic signal responsive to said violation predicted signal in order to signal traffic approaching said second traffic signal not to enter said intersection; said processing unit further operative to, maintain a virtual violation line, derive a position of said at least one vehicle from at least one of said plurality of images, compare the position of said vehicle to said virtual violation line, and generate a confirmation signal indicative of a red light violation in response to a determination that said at least one vehicle has crossed said virtual violation line during said red light phase of said first traffic signal. 30. The system of claim 29, wherein said violation capturing device comprises at least one video camera.31. The system of claim 29, wherein said violation image capturing device comprises at least one digital camera.32. The system of claim 29, wherein said collision avoidance unit comprises software executing on a processor.33. The system of claim 29, wherein said processing unit comprises software executing on a processor.34. The system of claim 29, wherein said processing unit is responsive to vehicle locations provided by a tracker unit.35. The system of claim 29, wherein said processing unit is further responsive to a time remaining in yellow light phase input.36. The system of claim 29, wherein said processing unit is further operable to determine a current speed for said at least one vehicle.37. The system of claim 29, wherein said processing unit is further operable to determine a current acceleration for said at least one vehicle.38. The system of claim 29, wherein said processing unit is further operable to compute a time remaining before one of said at least one vehicle enters said traffic intersection, responsive to determination of a current acceleration of said vehicle.39. The system of claim 38, wherein said processing unit is further operable to calculate a deceleration required for said at least one vehicle to stop within said time remaining before said vehicle enters said traffic intersection.40. The system of claim 39 wherein said processing unit further operative to determine whether said required deceleration is larger than a specified deceleration limit value, and if so, updates a violation prediction value for the current frame to indicate that a violation is predicted.41. The system of claim 29, wherein said virtual violation line is maintained by said processing unit as a portion of an internal representation of said intersection.42. The system of claim 41, wherein said virtual violation line is represented within said processing unit as being located beyond an actual stop line within a respective lane of said internal representation of said intersection.43. The system of claim 29, wherein said traffic light controller is responsive to said violation prediction to extending said red light phase of said second traffic signal.44. A method of collision avoidance at an intersection for a first traffic signal having a current light phase equal to one of the set consisting of at least red and green and a second traffic signal having a current light phase equal to one of the set consisting of at least red and green, comprising:capturing a plurality of images showing at least one vehicle approaching said first traffic signal, said images derived from an output of a violation image capturing device; maintaining at least one virtual violation line at an intersection for said at least one vehicle approaching said first traffic signal; generating, responsive to said plurality of images and an indication of said current first traffic signal light phase, at least one violation prediction for said at least one vehicle approaching said first traffic signal, said violation prediction indicating a likelihood that said at least one vehicle approaching said first traffic signal will violate an upcoming red light phase of said first traffic signal; asserting, responsive to said violation prediction, at least one violation predicted signal coupled to said second traffic signal; controlling, responsive to said violation predicted signal, said second traffic signal in order to signal traffic approaching said second traffic signal not to enter said intersection; generating from at least one of said plurality of images a location of said at least one vehicle with respect to said virtual violation line; comparing said location of said at least one vehicle to a position defined by said virtual violation line; and generating an output indicative of a red light violation of said first traffic signal in the event said location of said vehicle is determined to be beyond said position of said virtual violation line within said intersection during said red light phase of said first traffic signal. 45. The method of claim 44, wherein said violation image capturing device comprises at least one video camera.46. The method of claim 44, wherein said violation image capturing device comprises at least one digital camera.47. The method of claim 44, further including the steps of:determining a time remaining in a current yellow light phase; and generating said at least one violation prediction in response to said time remaining in said current yellow light phase. 48. The method of claim 44, further including the steps of:determining a current speed for said at least one vehicle; and generating said at least one violation prediction in response to said current speed of said at least one vehicle. 49. The method of claim 44, wherein said step of generating said at least one violation prediction includes the step of determining a current acceleration for said at least one vehicle.50. The method of claim 44, wherein said step of generating said at least one violation prediction includes the step of computing a time remaining before said at least one vehicle enters said traffic intersection.51. The method of claim 50, wherein said step of generating said at least one violation prediction further includes the step of calculating a rate of deceleration required for said at least one vehicle to stop within said time remaining before said vehicle enters said traffic intersection.52. The method of claim 51 wherein said step of generating said at least one violation prediction includes the step of determining whether said required deceleration is larger than a specified deceleration limit value, and if so, updating a violation prediction value for the current frame to indicate that a violation is predicted.53. The method of claim 44, wherein said step of maintaining said virtual violation line includes the step of maintaining said virtual violation line as a portion of a representation of said intersection.54. The method of claim 53, further including the step of maintaining said virtual violation line at a location beyond an actual stop line within a respective lane of said representation of said intersection.55. A method of avoiding collisions at an intersection, comprising:receiving data defining a virtual violation line from a user, the virtual violation line corresponding to a location at said intersection; storing a representation of said intersection and said virtual violation line; capturing images of a vehicle approaching said traffic signal at said intersection; analyzing said images to determine whether said vehicle is likely, during an upcoming red light phase of said traffic signal, to cross said virtual violation line; and upon determining that said vehicle is likely to cross said virtual violation line during said upcoming red light phase of said traffic signal, generating a signal operative to control an indicator to warn cross traffic approaching said intersection not to enter said intersection. 56. The method of claim 55 wherein the indicator-controlling signal indicates that a green phase of a traffic signal for said cross traffic is to be delayed.57. The method of claim 55, wherein said capturing step comprises capturing said images with a video camera.58. The method of claim 55, wherein said capturing step comprises capturing said images with a digital camera.59. The method of claim 55, wherein said analyzing includes determining a time remaining for said vehicle in a yellow light phase of said traffic signal, and the generating step is based in part upon said time remaining in said current yellow light phase.60. The method of claim 55, wherein generating said signal includes the steps of:determining from said plurality of images a current speed for said vehicle; and generating said signal based in part upon said current speed of said vehicle. 61. The method of claim 55, wherein the generating step includes:determining from said images a current acceleration for said vehicle; and generating said signal based in part upon said current acceleration of said vehicle. 62. The method of claim 55, wherein the generating step includes:generating a time remaining before said vehicle crosses said location corresponding to said virtual violation line; and generating said signal based in part upon said time remaining. 63. The method of claim 55, wherein the generating step includes:calculating a rate of deceleration that is required for said vehicle to stop before said location corresponding to said virtual violation line; and generating said signal in the event said required rate of deceleration is greater than a predetermined deceleration limit value. 64. The method of claim 55, wherein said representation of said virtual violation line corresponds to a location beyond an actual stop line at said intersection for a lane of traffic in which said vehicle is travelling.65. The method of claim 55, wherein said storing step includes storing said representation of said virtual violation line for locations corresponding to each of a plurality of lanes of traffic approaching said intersection.66. The method of claim 65 further including identifying which one of said plurality of lanes of traffic said vehicle is travelling in from an analysis of said images.67. The method of claim 65, wherein the user data defines the virtual violation line as being angled such that it is not crossed by a vehicle turning from one of the lanes onto a cross street.68. The method of claim 67, wherein the virtual violation line is defined as being angled such that it is not crossed by a vehicle turning right onto the cross street from a right-most one of the lanes.69. The method of claim 55, wherein the user data defines the virtual violation line as being parallel to an actual stop line at said intersection.70. A system for avoiding collisions at an intersection, comprising:a virtual violation line interface for receiving data defining a virtual violation line from a user, the virtual violation line corresponding to a location at said intersection; a storage device for storing a representation of said intersection and said virtual violation line; at least one camera for capturing images of a vehicle approaching said traffic signal at said intersection; and a processing unit operative: (1) to analyze said images to determine whether said vehicle is likely, during an upcoming red light phase of said traffic signal, to cross said virtual violation line, and (2) upon determining that said vehicle is likely to cross said virtual violation line during said upcoming red light phase of said traffic signal, to generate a signal operative to control an indicator to warm cross traffic approaching said intersection not to enter said intersection. 71. The system of claim 70, wherein the indicator-controlling signal indicates that a green phase of a traffic signal for said cross traffic is to be delayed.72. The system of claim 70, wherein said camera is a video camera.73. The system of claim 70, wherein said capturing step comprises capturing said images with a digital camera.74. The system of claim 70, wherein said processing unit is operable when analyzing said images to determine a time remaining for said vehicle in a yellow light phase of said traffic signal, and to generate said signal based in part upon said time remaining in said current yellow light phase.75. The system of claim 70, wherein said processing unit is operable when generating said signal to:determine from said images a current speed for said vehicle; and generate said signal based in part upon said current speed of said vehicle. 76. The system of claim 70, wherein said processing unit is operable when generating said signal to:determine from said images a current acceleration for said vehicle; and generate said signal is based in part upon said current acceleration of said vehicle. 77. The system of claim 70, wherein said processing unit is operable when generating said signal to:generate a time remaining before said vehicle crosses said location corresponding to said virtual violation line; and generate said signal based in part upon said time remaining. 78. The system of claim 70, wherein said processing unit is operable when generating said signal to:calculate a rate of deceleration that is required for said vehicle to stop before said location corresponding to said virtual violation line; and generate said signal in the event said required rate of deceleration is greater than a predetermined deceleration limit value. 79. The system of claim 70, wherein said representation of said virtual violation line stored in said storage device corresponds to a location beyond an actual stop line at said intersection for a lane of traffic in which said vehicle is travelling.80. The system of claim 70, wherein said storage device is operable to store said representation of said virtual violation line for locations corresponding to each of a plurality of lanes of traffic approaching said intersection.81. The system of claim 80, wherein said processing unit is operable to identify which one of said plurality of lanes of traffic said vehicle is travelling in from an analysis of at least one of said images.82. The system of claim 80, wherein the user data defines the virtual violation line as being angled such that it is not crossed by a vehicle turning from one of the lanes onto a cross street.83. The system of claim 82, wherein the virtual violation line is defined as being angled such that it is not crossed by a vehicle turning right onto the cross street from a right-most one of the lanes.84. The system of claim 80, wherein the user data defines the virtual violation line as being parallel to an actual stop line at said intersection.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.