System for detecting vehicles

• US 7,982,634 B2
• Filed: 03/19/2007
• Issued: 07/19/2011
• Est. Priority Date: 03/22/2006
• Status: Active Grant

• Current Assignee: Kria S.R.L.
• Original Assignee: Kria S.R.L.
• Inventors: Arrighetti, Stefano
• Primary Examiner(s): Crosland; Donnie
• Application Number:
  US12/293,947
• Publication Number:
  US 20090207046A1
• Time in Patent Office: 1,583 Days
• Field of Search: 340/937, 348/149, 382/105
• US Class Current: 340/937
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names
• CPC Class Codes: G08G 1/0175 by photographing vehicles, e.g. when violating traffic rules
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors
• CPC Class Codes: G08G 1/054 photographing overspeeding vehicles

Abstract

A system to detect the transit of vehicles having license plates includes at least one video camera to detect license plates capable of framing the plates of said vehicles and, preferably, at least one video camera to detect vehicles capable of framing a zone of transit of said vehicles having license plates. A series of processing operations is capable, starting from the video signals generated by the video camera to detect license plates, of detecting the presence of a vehicle in transit and, starting from the video signals generated by the video camera to detect vehicles, of detecting the position and three-dimensional shape of vehicles in transit in said zone. A supervisor module aggregates the results of these processing operations to generate records of information each identifying the modality of transit in said zone of a vehicle identified by a given license plate that has been recognized.

Claims

1. A system to detect the transit of vehicles having license plates, the system comprising:

at least one video camera to detect license plates capable of framing the license plates of said vehicles and of generating a corresponding video signal, and

a first processing chain sensitive to said video signal to perform license-plate-recognition processing on said video signal,

at least one video camera to detect said vehicles capable of framing a zone of transit of said vehicles having license plates and of generating a respective corresponding video signal, and

a second processing chain sensitive to said respective video signal to perform on said respective video signal vehicle-tracking processing to detect the position and three-dimensional shape of said vehicles in transit in said zone,

said first processing chain including a test module to detect the presence of a vehicle in transit by performing a test as to whether an image of a license plate persists on a series of images of said video signal, and a plate-tracking module which is activated by said test module when the license plate persists on the series of images,

a supervisor module that aggregates the results of said first processing chain and said second processing chain to generate information records, each record of said records identifying the transit modality in said zone of a vehicle identified by a given license plate tracked and recognized by said first processing chain.

2. The system according to claim 1, further comprising:

at least one video camera to detect vehicles capable of framing a zone of transit of said vehicles having license plates and of generating a respective corresponding video signal, and

a respective chain of processing sensitive to said respective video signal to perform on said respective video signal vehicle-tracking processing to detect the position and three-dimensional shape of vehicles in transit in said zone.

3. The system according to claim 2, further comprising a supervisor module that aggregates the results of said license-plate-recognition processing and said vehicle-tracking processing to generate information records, each of said records identifying the transit modality in said zone of a vehicle identified by a given license plate recognised by said chain of processing.

4. The system according to claim 2 wherein said video camera to detect license plates and said video camera to detect vehicles are included in a single video camera detection device.

5. The system according to claim 2 wherein said video camera to detect license plates and said video camera to detect vehicles are distinct co-positioned devices.

6. The system according to claim 1, wherein said video camera to detect license plates operates in the infrared spectrum.

7. The system according to claim 1, wherein said video camera to detect license plates is fitted with a narrow-field lens.

8. The system according to claim 1, wherein said video camera to detect license plates is provided with an illuminator that emits with wavelength in the same range of sensitivity of said video camera to detect license plates.

9. The system according to claim 8, wherein said illuminator is a stroboscopic illuminator and is synchronised with said video camera to detect license plates.

10. The system according to claim 2, wherein said video camera to detect vehicles is a video camera operating in the visible field.

11. The system according to claim 2, wherein said video camera to detect vehicles is a wide-field lens that frames all of said zone of transit of said vehicles having license plates.

12. The system according to claim 2, wherein said video camera to detect vehicles is a colour video camera, with a consequent possibility of detecting the condition of at least one light of adjacent traffic lights.

13. The system according to claim 1, wherein said video camera to detect license plates and/or said video camera to detect vehicles generate video signals provided with time/date imprint that identifies the instant at which they were recorded.

14. The system according to claim 1, further comprising:

said video camera to detect license plates being coupled with a homologous video camera to detect license plates to generate a corresponding homologous video signal, and

said chain of processing being sensitive to said video signal and to said homologous video signal to conduct on said video signal and on said homologous video signal stereometric processing to determine the speed of said vehicle in transit.

15. The system according to claim 2, wherein said respective chain of processing sensitive to said respective video signal is configured to perform on said respective video signal processing of the Structure From Motion or SFM type.

16. The system according to claim 2, further comprising:

said video camera to detect vehicles being coupled to a homologous video camera to detect vehicles and to generate a respective homologous video signal, and

said respective chain of processing being sensitive to said respective video signal and to said respective homologous video signal to perform stereometric processing on said respective video signal and on said respective homologous video signal to provide absolute characterisation of the motion and of the shape of said vehicle in transit.

17. The system according to claim 3, wherein said supervisor module is configured to sub-sample in a manner that is inversely proportional to the speed of transit of the vehicle, the sequences of images generated by said video camera to detect license plates and by said video camera to detect vehicles.

18. The system according to claim 3, wherein said supervisor module is configured to super-impress on the images of said vehicles in transit a vectorial representation with lines whose length is proportional to the velocity detected and whose direction identifies the trajectory of movement of the vehicle.

19. The system of claim 1, wherein said at least one detection video camera is constrained to a support through spacers of elastic material, and is protected by a housing without constraints to said housing.

20. The system according to claim 1, further comprising a pair of detecting video cameras, said chain of processing being capable of automatically detecting changes in the relative position between the video cameras of said pair.

21. The system according to claim 20, wherein said chain of processing is configured to automatically detect changes in the relative position between the video cameras of said pair detecting the failure to comply with an epipolar constraint in the correspondence between points in the respective pairs of images.