WO2017038129A1 - 違反者検出装置およびこれを備えた違反者検出システム - Google Patents

違反者検出装置およびこれを備えた違反者検出システム Download PDF

Info

Publication number
WO2017038129A1
WO2017038129A1 PCT/JP2016/057386 JP2016057386W WO2017038129A1 WO 2017038129 A1 WO2017038129 A1 WO 2017038129A1 JP 2016057386 W JP2016057386 W JP 2016057386W WO 2017038129 A1 WO2017038129 A1 WO 2017038129A1
Authority
WO
WIPO (PCT)
Prior art keywords
violation
image
violator
addict
detection
Prior art date
Application number
PCT/JP2016/057386
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
谷口　正宏
浩輝 上野
光司 滝沢
Original Assignee
オムロン株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by オムロン株式会社 filed Critical オムロン株式会社
Priority to CN201680036111.5A priority Critical patent/CN107710298B/zh
Publication of WO2017038129A1 publication Critical patent/WO2017038129A1/ja

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions

Definitions

  • the present invention relates to a violator detection device for detecting a violator who regularly repeats traffic violations such as signal ignorance and a violator detection system provided with the same.
  • Patent Document 1 in order to suppress pedestrians crossing the road ignoring such a red light, a violation person who entered the pedestrian crossing at the time of red light display is detected and installed in a traffic light.
  • An apparatus for displaying an image of the offender on a displayed display is disclosed.
  • the above-described conventional apparatus is intended to suppress violations by violators who have entered the pedestrian crossing while ignoring the red light, and does not consider detection of violators.
  • some traffic violators such as ignoring traffic lights are regularly repeated. Therefore, in order to suppress the occurrence of traffic accidents and traffic jams, it is important to take measures against these violators.
  • An object of the present invention is to provide a violator detection device capable of detecting a violator who regularly repeats traffic violations and a violator detection system including the same.
  • the offender detection device according to the first invention is a offender detection device for detecting offenders who regularly repeat traffic offenses among offenders including pedestrians or bicycle drivers who have committed traffic offenses.
  • storage part, and the addict detection part are provided.
  • the violation detection unit detects a traffic violation of a pedestrian or a bicycle driver.
  • An image acquisition part acquires the image which image
  • the addict detection unit detects the violator as a violation addict when the same person is shown as a violator in a plurality of images acquired by the image acquisition unit and photographed at different dates and times. To do.
  • the violation addict who repeats the traffic violation regularly is detected using the image image
  • the detection of the traffic violation in the violation detecting unit may be performed using, for example, image information captured by a camera or the like, or a sensor that detects an approach to a pedestrian crossing such as a pedestrian or a bicycle. You may perform using a detection result.
  • the image acquired in an image acquisition part is supplied from imaging devices, such as a camera attached to the road with a pedestrian crossing where the signal apparatus was installed, for example.
  • the violation addict is detected by detecting that there is a violator who is recognized as the same person in images taken at different dates and times acquired by the image acquisition unit.
  • the violator detection device determines whether or not the violator is the same person based on, for example, the violator's face, clothes, and body shape in the image. That is, in this violation detection device, when a traffic violation such as signal ignorance was detected, it was taken when an image of the violation and a past traffic violation stored in the storage unit were detected. Compare the image. When the violator included in the past image includes a violator who seems to be the same person, the violator detection device detects the violator as a violation addict.
  • the violator detection device is the violator detection device according to the first invention, wherein the addict detection unit includes the violator's face, gender, height, physique, estimated age,
  • a violation addict is detected by comparing a plurality of images using at least one of information on the type of riding bicycle, characteristic clothing, and belongings.
  • the offender who has committed traffic violations such as ignoring traffic lights based on the face, gender, height, physique, estimated age of the offender in the image, the type of bicycle on board, clothes, belongings, etc. Detect whether or not is a violation addict who has committed traffic violations in the past.
  • a violation addict can be detected using an image taken by a camera or the like installed near a pedestrian crossing.
  • the violator detection device is the violator detection device according to the first or second invention, and is based on the image of the violation addict detected by the addict detector and the detected image.
  • the information processing apparatus further includes a storage unit that stores the number of traffic violations of the violation addict and information on the violation date and time in association with each other.
  • the image of the violation addict is stored in the storage unit in association with information on the number of traffic violations and the date and time of violation (day of the violation, time zone, etc.).
  • the frequency of the violation addict committing the traffic violation for example, what time of day of the week the signal is ignored, etc. It is possible to grasp the tendency of addicted addicts, such as whether they are violating traffic.
  • the violator detection device is a violator detection device according to any one of the first to third inventions, among the images of violators addict detected in the addict detection unit, A display control unit that controls a display unit that displays at least one image is further provided.
  • the detection result in the addict detection unit that is, the image information related to the violation addict is displayed on a display unit such as a liquid crystal display.
  • the violator detection device is the violator detection device according to any one of the fourth invention, wherein the display control unit is detected by the addict detection unit together with the image of the violator addict.
  • the display section displays information including the number of violations of the violation addict based on the image, the day of the week with frequent violations, and the time zone.
  • information including the number of violations, days of the week with frequent violations, time zones, and the like is displayed on the display unit. This allows police officers, etc., who control traffic violations to conduct control at the site where traffic violations are detected on the day of the week or when the traffic violators frequently commit traffic violations such as ignoring traffic lights. .
  • a violator detection device is the violator detection device according to any one of the first to fifth aspects of the invention, wherein the violation detection unit includes an image acquired by the image acquisition unit, and an image The presence or absence of traffic violation of ignoring the signal is detected using at least one of the control information of the traffic lights included in the traffic light.
  • At least one of the information on the image acquired by the image acquisition unit and the control information of the traffic light included in the image is used.
  • an image including a pedestrian crossing a pedestrian crossing of a red signal an image including a pedestrian crossing a pedestrian crossing photographed at a time when the signal was red with the control information of the traffic light can be used to detect traffic violations such as ignoring traffic lights.
  • the violator detection device is the violator detection device according to the sixth invention, wherein the image acquisition unit acquires an image including the violator and a traffic light.
  • the above-mentioned violation addict is detected using an image including the violator and a traffic light as an image taken by an imaging device such as a camera. Thereby, the detection and verification of a traffic violation and the detection of a violation addict can be performed using the same image.
  • the violator detection device is the violator detection device according to the third invention, in the addict detection unit from a plurality of images stored in the storage unit and photographed at the same time
  • An image extraction unit is further provided for extracting an optimal image used for detection of a violation addict.
  • an optimal image used for detection of a violation addict is extracted based on a predetermined condition from a plurality of images acquired by the image acquisition unit and photographed at the same time.
  • the optimal image extraction conditions in the image extraction unit include, for example, the violator's face orientation that the violator's face is facing the camera, the violator's face clarity, image brightness, and the like.
  • the violation addict can be effectively controlled using the optimal image that satisfies the conditions necessary for detecting the violation addict.
  • the violator detection device is the violator detection device according to the eighth invention, wherein the direction of the face of the violator included in the image used for detection of the violator in the addict detection unit is determined.
  • a face orientation detection unit for detection is further provided.
  • the orientation of the violator's face included in the image acquired by the image acquisition unit is detected.
  • the optimal face orientation for detection of a violation addict for example, a state where the violator's face is facing the imaging device such as a camera can be considered. That is, the orientation of the violator's face in the image is detected by the face direction detection unit so that an image taken in a state where the violator's face stored in the storage unit is easily understood is extracted. As a result, an image taken with a face orientation that is easy to compare with the face of the offender in the image already stored in the storage unit can be extracted, and the offender can be detected efficiently.
  • the violator detection device is the violator detection device according to the eighth or ninth invention, wherein the image extraction unit is configured to detect the violator from a plurality of images stored in the storage unit. Extract the image when the traffic lights just before crossing the pedestrian crossing are confirmed.
  • an image when the violator included in the image acquired by the image acquisition unit confirms the color of the traffic light immediately before crossing the pedestrian crossing is extracted as an optimal image for detecting the violator.
  • pedestrians and bicycle drivers who cross the road ignoring the red light check the color of the traffic lights before crossing, check the traffic conditions of the road on the green light side, and then cross the road It is thought that.
  • the detection of a violation addict can be carried out more effectively by using an image taken with a pedestrian or the like looking up.
  • the violator detection device is the violator detection device according to the third invention, and the storage unit is an image when the violator ignores the signal and crosses the pedestrian crossing, and the violator is a traffic light.
  • the image when the color is confirmed is saved in association.
  • an image showing a situation in which a violation of red light is violated is associated with an image obtained by photographing a state where the violator confirms the color of the traffic light, and is stored in the storage unit.
  • the violation addict can be efficiently controlled by associating and storing the image for verifying the violation of the red light and the image for detecting the violation addict.
  • a violator detection device is the violator detection device according to any one of the first to eleventh aspects of the invention, wherein the addict detection unit has a plurality of images taken of the same place. Detect violation addicts from inside. Here, a violation addict is detected using a plurality of images taken at the same place. As a result, it is possible to detect a vicious violation addict who repeatedly repeats traffic violations at the same place and to carry out effective control.
  • a violator detection system is an violator detection device according to any one of the first to twelfth inventions and an image that is installed in the vicinity of a road including a pedestrian crossing and acquired by an image acquisition unit And an imaging device for photographing.
  • the violator detection system is configured using the above-described violator detection device and an imaging device that supplies an image including the violator to the violator detection device.
  • the imaging device is installed in the vicinity of a traffic signal installed on a pedestrian crossing, for example, so that an image in which the violator's face can be easily confirmed is obtained.
  • a traffic signal installed on a pedestrian crossing
  • an image in which the violator's face can be easily confirmed is obtained.
  • pedestrians and bicycles taken by a camera (imaging device) installed at an intersection with a crosswalk where a traffic light is installed, etc. are detected, and violation addicts are detected using the images. Can do.
  • traffic violation control can be carried out at the site where the traffic violation occurs based on the information of the violation addict detected by this system.
  • the violator detection system according to the fourteenth invention is the violator detection system according to the thirteenth invention, and the imaging device photographs a pedestrian or bicycle passing through a pedestrian crossing.
  • a pedestrian or bicycle passing through a pedestrian crossing where traffic violations such as red light are likely to occur is photographed by an imaging device such as a camera that supplies an image including the violator to the violator detection device.
  • an imaging device such as a camera that supplies an image including the violator to the violator detection device.
  • a violator detection system is the violator detection system according to the thirteenth or fourteenth invention, wherein the imaging device is installed in the vicinity of a pedestrian traffic light installed on a road including a pedestrian crossing.
  • an imaging device such as a camera that supplies an image including the violator to the violator detection device is placed in the vicinity of a pedestrian traffic light installed on a road including a pedestrian crossing that easily causes a violation of red light ignorance. It is installed.
  • the face of the offender who looked up at the traffic light to confirm the color of the traffic light just before crossing the road ignoring the red light can be photographed by an imaging device such as a camera.
  • an imaging device such as a camera
  • a violator detection system includes the violator detection device according to the fourth or fifth invention, and a display device for displaying a detection result in the addict detector.
  • a violator detection system including a display unit for displaying information on the addicted addict detected by the addict detection unit of the violator detection device is configured.
  • a display unit for displaying information on the addicted addict for example, a liquid crystal display, a PC (Personal Computer) monitor, or the like can be used.
  • the block diagram which shows the structure of the violator detection system containing the violator detection apparatus which concerns on one Embodiment of this invention.
  • the block diagram which shows the structure of the violation detection part contained in the violator detection apparatus of FIG. The block diagram which shows the structure of the image extraction part contained in the violator detection apparatus of FIG.
  • FIG. 11 is a diagram showing an image for confirming the face of a violation addict displayed on the display screen of FIGS. 8 to 10.
  • the flowchart which shows the flow of the violator detection process in the violator detection system of FIG.
  • the block diagram which shows the structure of the violation detection part contained in the violator detection apparatus of FIG. (A)-(c) is a figure which shows the example of installation of the camera attached to the traffic signal in the violator detection system which concerns on further another embodiment of this invention.
  • the violator detection apparatus 10 and the violator detection system 1 having the same according to an embodiment of the present invention will be described below with reference to FIGS. ⁇ 1, Configuration of the violator detection system 1>
  • the violator detection system 1 according to the present embodiment generates traffic violations such as signal ignorance by pedestrians, bicycles, etc. using images taken by a camera (imaging device) 20 installed at an intersection or a straight road. And violating addicts who regularly conduct traffic violations. For this reason, the violator detection system 1 is installed, for example, in a police station or the like that controls traffic violations.
  • the violator detection system 1 is used as a system for supporting police officers and the like so that they can efficiently control traffic violations.
  • a police officer or the like checks the display screen using a tablet terminal or a PC (Personal Computer) connected to the violator detection system 1.
  • the violator detection system 1 of the present embodiment includes a violator detection device 10, a camera (imaging device) 20, and a display unit 21.
  • the violator detection device 10 refers to an image taken by the camera 20 and image information of a past violator, and a pedestrian or a bicycle driver who has committed a traffic violation such as red light neglect. Detect violation addicts who regularly commit traffic violations.
  • the detailed configuration of the violator detection device 10 will be described in detail later. Further, in the present embodiment, a description will be given using a pedestrian traffic light 40 in which a red signal 40a is arranged in the upper stage and a display of the blue signal 40b is arranged in the lower stage. In the following description, it is assumed that all the traffic lights 40 included in the image indicate a state in which the red signal 40a side is lit.
  • the camera 20 that supplies an image to the violator detection device 10 is attached in the vicinity of the pedestrian traffic light 40 (here, the upper portion of the traffic light 40). Ignore the pedestrian A1 crossing the pedestrian crossing Z and photograph it from the front. At this time, the camera 20 shoots an image including the pedestrian A1 crossing the pedestrian crossing Z ignoring the red signal 40a and the traffic light 40 next to or behind the pedestrian A1 (the red signal 40a is lit). It is installed as follows.
  • the image P1 (refer FIG. 9 etc.) including the traffic light 40 at the time of red signal 40a lighting with the pedestrian A1 who committed the violation can be acquired. Therefore, the image P1 can be used as an image for verifying a traffic violation ignoring a red signal and for detecting a violation addict.
  • the camera 20 also takes a picture of the driver A2 of the bicycle B crossing the pedestrian crossing Z ignoring the red signal 40a from the front as shown in FIG.
  • the camera 20 continuously captures images including the pedestrian crossing Z and the pedestrian A1 who crosses the pedestrian crossing Z and the driver A2 of the bicycle B. Then, the camera 20 transmits the captured image to the image acquisition unit 11 of the violator detection system 1 described later.
  • the camera 20 continuously captures an image having a pixel number of 5M at 15 frames per second, for example.
  • the image captured by the camera 20 may be a still image captured continuously or a moving image.
  • photography with the camera 20 may be always performed continuously, and may be performed only for the predetermined time after detecting a traffic violation.
  • the camera 20 may perform shooting only while detecting the presence of a person in the shooting area regardless of whether there is a traffic violation.
  • the camera 20 is disposed so as to face the pedestrian crossing Z. For this reason, each camera 20 can photograph the pedestrian A1 and the driver A2 of the bicycle B who cross the pedestrian crossing Z from the front.
  • the camera 20 is attached to the upper part of the traffic light 40, the camera 20 is installed at a position higher than the eyes of the pedestrian A1 and the driver A2 of the bicycle B. Is done.
  • the camera 20 can photograph the face of the pedestrian A1 or the like crossing the pedestrian crossing Z from a slightly upper front, the image of the violator's face is used as an image used when detecting an addicted addict. It is possible to obtain an optimal image that easily recognizes the feature.
  • the pedestrian A1 and the like who cross the pedestrian crossing Z ignoring the red signal 40a confirm the color of the traffic light 40 immediately before ignoring the signal.
  • the camera 20 is attached to the traffic light 40.
  • the image taken by the camera 20 when a violator such as a pedestrian A1 who wants to ignore the signal looks at the traffic light 40 is a state in which the violator looks at the camera 20 (the face of the violator is the camera 20). Therefore, since the violator's face is always photographed in a state of facing the camera 20, an image that can be easily compared with the past violators' images can be obtained. As a result, it is possible to further improve the accuracy of detecting a violation addict as compared to the case of using an image obtained by normally photographing a pedestrian A1 crossing the pedestrian crossing Z.
  • the display unit 21 included in the violator detection system 1 of the present embodiment is a display device such as a liquid crystal display, for example.
  • the display unit 21 displays contents (types) of traffic violations detected, various images, and the like when checking traffic violation violation addicts such as ignoring traffic signals from pedestrians and bicycles.
  • the display part 21 is the information regarding the traffic violation and violation addict who were acquired in the image acquisition part 11 by the display control part 16 of the violator detection apparatus 10 mentioned later, and the 1 image or several image linked
  • the display mode on the display screen of the display unit 21 will be described in detail later.
  • the violator detection device 10 of the present embodiment acquires an image taken by the camera 20 and detects a traffic violation that crosses the pedestrian crossing Z ignoring the red signal 40a. To do. Furthermore, the violator detection device 10 detects a violation addict (a pedestrian A1 or a driver A2 of a bicycle B) who regularly violates violations. That is, the violator detection device 10 is used to detect traffic violations such as ignoring red traffic lights and to control the violators on the spot.
  • a violation addict a pedestrian A1 or a driver A2 of a bicycle B
  • the violator detection device 10 includes an image acquisition unit 11, a violation detection unit 12, an image extraction unit 13, a violation addict detection unit 14, a storage unit 15, Display control unit 16.
  • the image acquisition unit 11 acquires a plurality of images continuously captured by the camera 20. Then, the image acquisition unit 11 transmits the plurality of images to the violation detection unit 12 and the image extraction unit 13.
  • the violation detection unit 12 uses the plurality of images received from the image acquisition unit 11 to select an image that satisfies a necessary condition for verifying a specific traffic violation such as signal ignorance. Specifically, the violation detection unit 12 selects an image including the pedestrian A1 and the like and the traffic light 40 crossing the pedestrian crossing Z ignoring the red signal 40a from a plurality of images. When this image is selected, an optimal image or a plurality of images are extracted based on conditions necessary for verifying traffic violations such as image appearance, content, and clarity.
  • the violation detection unit 12 determines whether the red signal 40a is detected based on the switching information (signal control signal) of the traffic light 40 or the display of the red signal 40a detected from the image. It can be determined whether or not the signal 40a is lit. And the violation detection part 12 detects the offender who crosses the pedestrian crossing Z at the time of lighting of the red signal 40a. More specifically, for example, in one image, when the traffic light 40 in the image is a red signal 40a, an image in which a pedestrian A1 or a driver A2 of a bicycle B enters the pedestrian crossing Z is used. Thus, it is possible to detect traffic violations ignoring traffic lights (see FIG. 8 etc.).
  • the image extraction unit 13 extracts an optimum image from the image information received from the image acquisition unit 11 in order to detect traffic violation violation addicts such as signal ignorance detected by the violation detection unit 12.
  • the optimum image condition extracted by the image extraction unit 13 includes, for example, an image obtained by photographing the violator's face from the front.
  • the image extraction unit 13 includes a face orientation detection unit 13a as shown in FIG.
  • the face orientation detection unit 13a detects the orientations of the faces of the pedestrian A1 and the driver A2 of the bicycle B included in the plurality of images received from the image acquisition unit 11. Thereby, the image extraction part 13 can extract the image in the state where the face direction of the pedestrian A1 or the like faces the traffic signal 40 in which the camera 20 is installed in the face direction detection part 13a.
  • the violation addict detection unit 14 receives the optimum image extracted by the image extraction unit 13 and compares the image of the violator with the past image of the violator stored in the storage unit 15. If the same person appears as a violator in images taken at different dates and times, the person is detected as a violation addict who has repeatedly repeated traffic violations.
  • the violation addict detection unit 14 includes an optimal image including the violator received from the image extraction unit 13, an image of the violator who has committed a traffic violation in the past, and is already stored in the storage unit 15. Compare Then, it is determined whether or not the violators included in the image are the same person. This determination operation is performed on a plurality of images. Here, the determination as to whether or not they are the same person is made according to the facial features of the violator, body type, estimated age, sex, clothing, belongings, type of bicycle, color, etc.
  • the violation addict detection process in the violation addict detection unit 14 will be described in detail later with reference to FIG.
  • the storage unit 15 includes the contents of traffic violations such as signal ignorance detected by the violation detection unit 12, the optimum image for proof of violation extracted by the image extraction unit 13, and the violation addict detected by the violation addict detection unit 14. Information, images for detecting violators, etc. are stored in association with each other.
  • the storage unit 15 does not store images that are not related to traffic violations such as signal ignorance, unclear images that are not suitable for verification of traffic violations, images that are not used for detection of violation addicts, and the like.
  • the display control unit 16 effectively displays information on traffic violations by the addicted addicts stored in the storage unit 15, information on the addicted addicts, and one or more images associated therewith.
  • the display unit 21 is controlled.
  • the violation detection unit 12 includes a DB (database) 12a and a determination unit 12b.
  • the DB 12a stores information on conditions (including image conditions that can be used for verification) for verifying traffic violations such as signal ignorance.
  • the conditions for verifying a traffic violation such as signal ignorance are image information including the traffic signal 40 on the road with the pedestrian crossing Z where the camera 20 is installed, information on switching of the traffic signal 40 (signal control signal, etc.) Etc. are included.
  • the image condition that can be used for verification means a condition that the image content as evidence must satisfy the detected traffic violation.
  • conditions such as a red signal 40a and a pedestrian A1 who is a violator are included in one image.
  • the first image includes the pedestrian A1 who has entered the pedestrian crossing Z ignoring the red signal 40a, and the second image
  • there are conditions such as including a face portion when the pedestrian A1 is photographed from the front.
  • the determination unit 12 b collates images extracted from a plurality of images acquired by the image acquisition unit 11 based on the verification conditions stored in the DB 12 a and is included in the images. It detects whether pedestrian A1 etc. committed the traffic violation of signal ignorance. If it is determined that the traffic violation can be verified as ignoring the signal, the violation identification image P1 (see FIG. 8 and the like) suitable for the verification as the traffic violation is extracted and transmitted to the storage unit 15. To do.
  • a processing area for every function in an image of 5 million pixels, for example.
  • an area including a traffic light is set as a red signal detection processing area
  • an area from the vicinity of the stop line of the pedestrian crossing Z to the pedestrian crossing Z is set as a violator detection processing area.
  • the red signal detection processing area for example, by detecting the timing when the traffic light 40 changes to a red signal by image processing, it becomes a trigger when detecting the pedestrian A1 and the bicycle B that ignore the red signal.
  • the determination unit 12 b outputs the presence / absence of traffic violation and violation detection information (such as an image P ⁇ b> 1 for specifying violation, information on violation content) to the storage unit 15. Furthermore, the determination means 12b outputs violation detection information to the image extraction part 13, when a traffic violation is detected.
  • the image extraction unit 13 extracts an optimum image from the image information received from the image acquisition unit 11 in order to detect traffic violation violation addicts such as signal ignorance detected by the violation detection unit 12.
  • the image extraction unit 13 includes a face orientation detection unit 13a as illustrated in FIG. 5 in order to extract an optimal image.
  • the face direction detecting unit 13a includes a violator (pedestrian A1 or The direction of the face of the driver B2 of the bicycle B) is detected. More specifically, the face direction detection unit 13a detects the direction of the violator's face included in an image in which traffic violation such as signal ignorance is detected, and extracts an image in which the violator's face is photographed from the front. .
  • the image extraction unit 13 eliminates images that cannot be used for detection of violation addicts, such as an image in which the pedestrian A1 who committed signal ignorance is walking down or sideways, and is a violation addict. Only images that are optimal for detection are extracted. Then, the image extraction unit 13 outputs the extracted optimal image information to the violation addict detection unit 14 and the storage unit 15.
  • the violation addict detection unit 14 detects a violation addict using, for example, an image taken in the same direction as the past violator image, such as an image of the violator face taken from the front. be able to. Further, the storage unit 15 can store image information that is optimal for detection of a violation addict in association with violation content detected by the violation detection unit 12.
  • the optimum image information extracted by the image extraction unit 13 is not limited to an image obtained by photographing the offender's face from the front, but an image representing the characteristics of the offender, such as other images in which the profile and clothes are easy to understand. It may be.
  • the violation addict detection unit 14 uses the images taken at different dates and times by the camera 20 to detect a violation addict as follows. That is, as shown in FIG. 6, the violation addict detection unit 14 receives optimal image information indicating the characteristics (face, etc.) of the violator extracted by the image extraction unit 13 and stores it in the storage unit 15. Compare with saved image information of past violators. Specifically, the image information of the violator detected this time is compared with the image information of the violator detected in the past (different date and time) stored in the storage unit 15 to identify the same violator. Detects past images taken.
  • whether or not they are the same violators is determined with reference to the portions showing the characteristics of the person, such as the face, figure, gender, clothes, personal belongings, etc. included in the image of the violators.
  • the features of the violators in the image acquired this time and the past violators' faces can be efficiently obtained by graphicizing or digitizing the features by image processing etc. Can be compared.
  • the violation addict detection unit 14 determines whether or not the person has repeatedly violated a predetermined number of times. If it is determined that the violation has been repeated a predetermined number of times or more, this person is determined as a violation addict, and information on the violation addict is sent to the storage unit 15. The information on the violation addict sent to the storage unit 15 is stored as optimum image information so that it can be used for detection of a future violation addict. In addition, in the violation addict detection unit 14, the optimum image information of the violator determined not to be a violation addict is also used as the optimum image information so that it can be used for detection of a future violation addict. Saved.
  • the predetermined number of times used for determining whether or not the person is a violation addict for example, it may be five times or more, and in order to make a person who has violated once in the past a violation addict It may be two or more times. In other words, the number of predetermined times can be arbitrarily set in accordance with the traffic violation occurrence state or the like when the traffic violation such as ignoring the signal at the police station is controlled.
  • the storage unit 15 is a storage means such as a flash memory capable of writing and erasing data, and as shown in FIG. 7, an image information storage area 15a, a violation detection information storage area 15b, and a violation addict information storage area 15c. have.
  • the image information storage area 15a shows the image information acquired by the image acquisition unit 11 described above, the image information related to the traffic violation detected by the violation detection unit 12, and the characteristics of the violator extracted by the image extraction unit 13. Save optimal image information.
  • the image information storage area 15a includes -Image information including traffic lights and violators acquired in the image acquisition unit 11-Image information including faces of violators acquired in the image acquisition unit 11-For verifying traffic violations used in the violation detection unit 12
  • the image information and image information such as the image information of the violation addict detected by the violation addict detection unit 14 are stored.
  • the violation detection information storage area 15b stores information related to traffic violation contents (violation contents, violation location, violation date, etc.) such as signal ignorance detected by the violation detection unit 12.
  • the violation addict information storage area 15c stores the characteristics of the violator determined as a violation addict. More specifically, the violation addict information storage area 15c includes: ⁇ Violation addict's violation contents, number of violations, violation date and time (day and time when there are many violations) ⁇ Characteristics of violation addict (face, gender, physique, estimated age, etc.) ⁇ Characteristics of violation addicts (clothing, belongings, type of bicycle, color, etc.) Save various information.
  • the display control unit 16 controls the display on the display unit 21.
  • police officers or the like who control traffic violations such as signal ignorance can confirm the contents displayed on the display unit 21 and efficiently perform the control.
  • the display unit 21 displays violation contents regarding traffic violations such as signal ignorance, various information regarding violation addicts, and images P1 to P10.
  • FIG. 8 shows a state in which information on the offender detected as a violation addict is displayed on the display screen S ⁇ b> 1 of the display unit 21.
  • the display screen S ⁇ b> 1 is a screen showing the contents of one violation, and has three display areas: a violation addict switching area 51, a violation contents display area 52, and a violation image display area 53. is doing.
  • the violation addict switching area 51 two switching sections 51a and 51b are displayed.
  • the switching units 51a and 51b are switched to a screen that displays other addicts.
  • the number of the detected violation addict, the content of the violation (for example, ignoring red light), the violation location, the number of violations detected by the violation addict, and the shooting date and time of the display image are displayed.
  • the traffic violation is verified and the violation addict is detected using two images.
  • the violation addict displayed in the violation content display area 52 is displayed not as personal information such as a name but as a number.
  • the reason for this is that the violator detection system 1 is not intended to identify the personal information of the addicted addict, but is a police officer who detects the addicted addict and controls the traffic violation. This is because it is intended to be provided to the public. Therefore, in this violator detection system, it is not necessary to identify the violator addict individual by referring to personal information by face authentication or the like using the acquired image.
  • the violation image display area 53 displays the violation content displayed in the violation content display area 52 and an image for detecting a violation addict.
  • one violation detection image P1 and one face confirmation image P2 are displayed.
  • the violation detection image P ⁇ b> 1 indicates an image captured so as to include the red traffic light 40 and the pedestrian A ⁇ b> 1 entering the pedestrian crossing Z.
  • An image P1 obtained by photographing such a violation addict from the front can be easily obtained because the camera 20 is attached to the traffic light 40 installed in the pedestrian crossing Z or in the vicinity thereof. Thereby, the traffic violation of ignoring the red signal can be verified by the single image P1 including the traffic light 40 of the red signal and the pedestrian A1 in the pedestrian crossing Z.
  • the image P2 for checking the face of the offender displays the part of the face of the offender who has been determined to ignore the signal in the offending detection image P1.
  • the image P2 may be an image obtained by enlarging the face portion of the violation detection image P1, or may be an image obtained by enlarging another image, or an image obtained by enlarging the face portion from the beginning. May be used.
  • the face of the offender can be enlarged and confirmed. For this reason, it is possible to obtain an image that can be easily compared with the face included in the image information of past violators stored in the storage unit 15 in advance.
  • the violation image display area 53 is provided with image switching units 53a and 53b on the left and right sides of the two pieces of image information. When an operation such as touch is performed on the tablet terminal, the image switching units 53a and 53b can switch the display to an image other than the images P1 and P2 displayed on the current display screen S1.
  • FIG. 9 shows a display when displaying information related to one addicted person on one display screen S2.
  • the display screen S ⁇ b> 2 has three display areas, a violation addict switching area 51, a violation content display area 52, and a violation image display area 53, as shown in FIG. 9. .
  • the display screen S2 is different from the display screen S1 of FIG. 8 described above in that the violation image display area 53 displays images P3 to P6 of the same violator taken at different dates and times. .
  • the display content other than the violation image display area 53 is the same as that of the display screen S1, detailed description thereof is omitted here.
  • the violation image display area 53 image information of violation addicts photographed at different dates and times is displayed. Specifically, the violating image display area 53 displays images P3 to P6 of the violators detected as the same person, which were taken at four dates and times of shooting dates A, B, C, and D, respectively.
  • a police officer or the like can see the image of the violation addict determined as the same person, and can recognize the features of the violation addict such as the face, body shape, sex, and clothes. Furthermore, when the police officers, etc., see the display screen S2 and a similar but different person is displayed among the images of the violation addicts detected as the same person, the person is designated and excluded. can do. It should be noted that images of other violators excluded here are assigned different addicted addict numbers and are stored again in the storage unit 15.
  • FIG. 10 shows a display screen for various information related to a plurality of addicted addicts regarding traffic violations of signal ignorance displayed on the display screen S3 of the display unit 21.
  • FIG. 10 shows an example in which a plurality of violation addicts detected at the same violation location are displayed side by side.
  • the display screen S3 includes, in addition to the two display areas, the violation content display area 52 and the violation image display area 53, as shown in FIG. A page switching area 54 is provided.
  • the display content in the violation content display area 52 is the same as that of the display screens S1 and S2 described above, and thus detailed description thereof is omitted here.
  • the violation point switching area 55 two switching parts 55a and 55b are displayed.
  • the switching units 55a and 55b are switched to a screen that displays violation addicts detected at other points.
  • the violation image display area 53 of the display screen S3 displays images P7 to P10 of four violation addicts detected as violation addicts among traffic violation violators ignoring red traffic lights.
  • the violations image display area 53 of the display screen S3 the images P7 to P10 of violation addicts with different violation addict numbers are displayed.
  • the violation content display area 52 information on the number assigned to each violation addict, the number of violation detections, the shooting date and time, and the like is displayed.
  • this display screen S3 when an addicted addict who appears to be the same person is displayed by looking at the facial features, body shape, clothes, etc. of the person in the image, the image information and the violation information are collected together. You can also re-register as a violation addict. In this case, the number of the violation addict is unified to the old number, the new number is deleted, and the number of violation detections may be displayed again by adding the number corresponding to each number.
  • a page switching button 54a for switching display to the previous page and a page switching button 54b for switching display to the next page are displayed.
  • these page switching buttons 54a and 54b are touched on the display screen of the tablet terminal, the display screen is switched to the display screen before or after the display screen.
  • the violator detection apparatus 10 of the present embodiment uses image information including the violator's face photographed in the past and stored in the storage unit 15 when detecting a violation addict.
  • the image information including the face is usually taken from the front of the face.
  • the violator images P2 to P10 for face confirmation displayed in the violating image display area 53 of the display screens S1 to S3 described above are formed as follows.
  • the images P2 to P10 of the violators for face confirmation include the pedestrian A1 who has seen the traffic light 40 immediately before crossing the red traffic light Z and the traffic light 40 on the side thereof.
  • the image P11 is formed by enlarging the part of the violator's face in the image P11.
  • the pedestrian A1 and the driver A2 of the bicycle B who cross the pedestrian crossing ignoring the red light confirm the color of the traffic light 40 immediately before crossing the pedestrian crossing.
  • the pedestrian A1 and the like who confirmed the color of the traffic light 40 are in a state of facing each other with their face up against the camera 20 attached to the traffic light 40 arranged oppositely.
  • the camera 20 can photograph the face portion of the pedestrian A1 or the like who has entered the pedestrian crossing Z while ignoring the red light by photographing the image at this moment. Therefore, images with the violator's face facing the camera 20 can be obtained as images for detecting a violation addict, that is, images P2 to P10 in which the face of the violator is enlarged for face confirmation. .
  • These images P2 to P10 are stored in the storage unit 15 as violation information in a state associated with the violation specifying image P1 and the violation addict specifying image P11.
  • step S11 the image acquisition unit 11 acquires an image including a pedestrian crossing where a traffic light is installed.
  • the image acquired here may be a continuously shot image or a moving image.
  • step S12 the violation detection unit 12 detects whether or not a traffic violation such as signal ignorance has occurred and the violator is detected in the acquired image.
  • a traffic violation and its violator are detected, the process proceeds to step S13.
  • the process returns to step S11.
  • step S ⁇ b> 13 the image extraction unit 13 extracts an optimum image for specifying the violation and confirming the face of the offender.
  • the optimum image extracted by the image extraction unit 13 is extracted according to the direction of the violator's face in the image detected by the face direction detection unit 13a, the sharpness of the image, and the like.
  • the violation addict detection unit 14 collates the optimal image for face confirmation extracted in step S ⁇ b> 13 with the optimal image of past violators already stored in the storage unit 15.
  • step S15 the violation addict detection unit 14 determines whether or not the same violator image is included in the violator images taken in the past (different dates and times). If it is determined that there is an image of the same violator, the process proceeds to step S16. On the other hand, if it is determined that there is no image of the same violator, the process proceeds to step S19. Next, after it is determined in step S16 that there is the same violator, the image for confirming the face of the violator and the image of the previous violator are associated and stored together with the violation date and the number of violations is updated. .
  • step S17 it is determined whether or not the number of violations updated in step S16 exceeds a predetermined number. If it is determined that the number of times is equal to or greater than the predetermined number, the process proceeds to step S18. On the other hand, if it is determined that the number is less than the predetermined number, the process returns to step S11.
  • step S18 if it is determined that the violator determined as a violation addict in step S17 is a violator more than a predetermined number of times, this violator is assigned a violation addict number as a violation addict. Is done.
  • the violation addict number is stored in the storage unit 15 by associating the information of the violation addict (the day of the week and the date / time when there are many violations) with various image information and the like.
  • step S15 if it is determined in step S15 that the same violator does not exist, a new violator number is assigned as the first violator in step S19.
  • the violator number is stored in association with violation information (ignoring signal, etc.), violation date and time, together with image information.
  • the violator detection system 1 and the violator detection device 10 of the present embodiment compare the image of the violator with the traffic violator and the image of the violator in the past. In this case, it is determined whether the violation is caused by the same person.
  • the violation when it is determined that the violation is caused by the same person, it is registered as a violation addict, and the content of the violation and the number of violations are stored together with the image information for detecting the violation and the image information indicating the characteristics of the offender. Save to.
  • the addicted addict after grasping the characteristics (faces, etc.) of the addicted addict, the addicted addict frequently commits traffic violations on the day of the week and at the time of day, and can wait on the spot.
  • a police officer or the like can crack down with a current offender when a person who appears to be a violation addict appears on the road and commits a traffic violation that ignores the signal.
  • the violation detection unit 12 receives a signal control signal and the detection result of the sensor 22 and detects a traffic violation such as ignoring a red signal.
  • the system 101 may be used.
  • the breach detection unit 112 detects a pedestrian or the like in the pedestrian crossing along with a signal control signal from the signal control device 30 that controls the traffic light 40. May be received.
  • the sensor 22 detects that a pedestrian, a bicycle, or the like has entered the pedestrian crossing during the red signal time zone that can be confirmed by the signal control signal, so that the determination means refers to the verification condition stored in the DB 112a. Red signal neglect can be detected at 112b.
  • the violation detection means the violation detection based on the image information shown in FIG. 1 and the combination of the sensor detection result and the signal control signal shown in FIG. 13 may be used in combination.
  • the violator may be detected as an addicted addict.
  • the violator instead of detecting the same person by comparing the violator who committed the traffic violation detected in real time with the image information of the already stored violator, it is the same among multiple violators already stored You may search for a person and detect a violation addict.
  • FIG. 15A a square camera 120 in a front view may be installed below the traffic light 40.
  • FIGS. 15B and 15C a camera 120 having a similar shape may be installed on the upper side or the side surface of the traffic signal 40.
  • an imaging device such as a camera may be installed on a support or the like different from the traffic light.
  • violators who ignore the signal will check the color of the traffic light before crossing the pedestrian crossing. For this reason, when taking an image of the moment when the violator confirms the color of the traffic light, inevitably the violator's face is directly facing the camera. More preferably, it is installed in the vicinity.
  • a violation addict may be detected by detecting other traffic violations such as bicycle two-seaters, traveling zone violations, translational traffic violations, temporary suspension violations, and the like.
  • a violation addict may be detected by detecting no light on the bicycle at night using an infrared camera or the like. That is, the present invention can be applied to detection of other types of traffic violations and violation addicts in accordance with the traffic rules defined by the country or region.
  • the violator detection device of the present invention has the effect of being able to detect violators who regularly repeat traffic violations such as ignoring red traffic lights, etc., for traffic management systems installed on main roads, etc. Widely applicable.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)
  • Closed-Circuit Television Systems (AREA)
PCT/JP2016/057386 2015-09-03 2016-03-09 違反者検出装置およびこれを備えた違反者検出システム WO2017038129A1 (ja)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201680036111.5A CN107710298B (zh) 2015-09-03 2016-03-09 违章者检测装置及具备其的违章者检测系统

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2015173916A JP6398920B2 (ja) 2015-09-03 2015-09-03 違反者検出装置およびこれを備えた違反者検出システム
JP2015-173916 2015-09-03

Publications (1)

Publication Number Publication Date
WO2017038129A1 true WO2017038129A1 (ja) 2017-03-09

Family

ID=58187035

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/057386 WO2017038129A1 (ja) 2015-09-03 2016-03-09 違反者検出装置およびこれを備えた違反者検出システム

Country Status (3)

Country Link
JP (1) JP6398920B2 (zh)
CN (1) CN107710298B (zh)
WO (1) WO2017038129A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021082505A1 (zh) * 2019-10-28 2021-05-06 深圳市商汤科技有限公司 图片处理方法、装置、设备、存储介质和计算机程序

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6634887B2 (ja) * 2016-03-02 2020-01-22 株式会社リコー 情報提供システム、方法、プログラムおよびネットワークシステム
CN107610449A (zh) * 2017-08-30 2018-01-19 顾泰来 一种交互式平台监督管理系统
CN107622658B (zh) * 2017-10-12 2018-10-12 樊陆 行人过马路的监测方法及终端设备
CN110223505A (zh) * 2018-03-02 2019-09-10 杭州海康威视数字技术股份有限公司 一种行人闯红灯行为的确定方法及装置
CN111832352A (zh) * 2019-04-18 2020-10-27 杭州海康威视系统技术有限公司 非机动车监控方法、装置及设备、存储介质
CN114973664A (zh) * 2022-05-17 2022-08-30 保银信科信息技术(湖北)有限公司 一种基于机器视觉的城市交通在线智能监测预警管理系统

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003346280A (ja) * 2002-05-29 2003-12-05 Fujitsu Ltd 交通信号機用警告装置
JP2005301377A (ja) * 2004-04-07 2005-10-27 Nec Corp 車両照会システム及びそれに用いる車両照会方法
JP2006252139A (ja) * 2005-03-10 2006-09-21 Omron Corp 車両照会システム

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4218348B2 (ja) * 2003-01-17 2009-02-04 オムロン株式会社 撮影装置
US20050156757A1 (en) * 2004-01-20 2005-07-21 Garner Michael L. Red light violation prevention and collision avoidance system
US8297977B2 (en) * 2005-07-12 2012-10-30 Eastern Virginia Medical School System and method for automatic driver evaluation
CN101261722A (zh) * 2008-01-17 2008-09-10 北京航空航天大学 电子警察后台智能管理和自动实施系统
CN102930721B (zh) * 2012-10-29 2014-11-05 张剑 交通违规监控与处罚系统
CN103021175A (zh) * 2012-11-12 2013-04-03 上海经达实业发展有限公司 基于达芬奇架构的行人闯红灯视频检测方法及装置
CN103077608B (zh) * 2012-12-07 2014-12-03 哈尔滨工业大学 一种取证驾驶员是否故意闯黄灯的方法
CN203325221U (zh) * 2013-03-26 2013-12-04 广东万安科技股份有限公司 一种多功能的电警摄像机
CN103236166B (zh) * 2013-05-13 2015-05-27 交通运输部公路科学研究所 一种使用卫星定位技术进行车辆违章行为辨识的方法
CN103366565B (zh) * 2013-06-21 2015-04-15 浙江理工大学 一种基于Kinect的行人闯红灯检测方法及其系统
CN103606278A (zh) * 2013-11-25 2014-02-26 潘保林 交通路口行人违规的综合管理装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003346280A (ja) * 2002-05-29 2003-12-05 Fujitsu Ltd 交通信号機用警告装置
JP2005301377A (ja) * 2004-04-07 2005-10-27 Nec Corp 車両照会システム及びそれに用いる車両照会方法
JP2006252139A (ja) * 2005-03-10 2006-09-21 Omron Corp 車両照会システム

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021082505A1 (zh) * 2019-10-28 2021-05-06 深圳市商汤科技有限公司 图片处理方法、装置、设备、存储介质和计算机程序

Also Published As

Publication number Publication date
JP2017049884A (ja) 2017-03-09
CN107710298B (zh) 2020-09-29
CN107710298A (zh) 2018-02-16
JP6398920B2 (ja) 2018-10-03

Similar Documents

Publication Publication Date Title
JP6398920B2 (ja) 違反者検出装置およびこれを備えた違反者検出システム
CN103065470B (zh) 基于单目多检测面的机器视觉的车辆闯红灯行为检测装置
JP6252693B2 (ja) 表示装置およびこれを備えた交通違反管理システム
US20180240336A1 (en) Multi-stream based traffic enforcement for complex scenarios
JP6394402B2 (ja) 交通違反管理システムおよび交通違反管理方法
WO2016113973A1 (ja) 交通違反管理システムおよび交通違反管理方法
CN101739809A (zh) 行人闯红灯自动报警监控系统
KR101143521B1 (ko) 교차로 통행위반 차량 단속 시스템 및 이를 이용한 교차로 통행위반 차량 단속방법
CN201307337Y (zh) 行人闯红灯自动报警监控装置
JP2006260483A (ja) 顔照合システム及び顔照合方法
KR101001302B1 (ko) 코너정보를 이용한 끼어들기 위반차량 검지시스템 및 그 방법
WO2017038128A1 (ja) 違反者特定装置およびこれを備えた違反者特定システム
WO2017038125A1 (ja) 違反抑止装置およびこれを備えた違反抑止システム
KR20160067435A (ko) 횡단보도의 무단횡단 알림 시스템
JP2003178397A (ja) 道路表示等検出警告装置
JP6364652B2 (ja) 交通違反管理システムおよび交通違反管理方法
KR102510765B1 (ko) 이륜차 단속 시스템 및 방법.
KR101394201B1 (ko) 버스장착 무선cctv 촬영단속시스템
JP2005234774A (ja) 交通信号無視車両警告装置、及び交通信号無視車両警告・記録装置
WO2017022268A1 (ja) 車両特定装置およびこれを備えた車両特定システム
EP2143092B1 (en) System and method for monitoring and capturing potential traffic infractions
KR200416652Y1 (ko) 주행형 불법 주정차 단속 시스템
KR101514408B1 (ko) 광고 수익을 이용한 횡단 보도 안전 시스템 및 방법
JP6398919B2 (ja) 違反抑止装置およびこれを備えた違反抑止システム
AU785266B2 (en) Traffic violation detection system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16841162

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16841162

Country of ref document: EP

Kind code of ref document: A1