The present disclosure provides a method and apparatus for detecting a traffic anomaly, a device, a storage medium and a computer program product, relates to the field of artificial intelligence, and specifically to computer vision and deep learning technologies, and can be applied to intelligent transportation scenarios. A specific implementation of the method comprises: acquiring a traffic video stream; performing vehicle detection tracking on the traffic video stream to determine whether there is an abnormally stopped vehicle, wherein a stop with a time length exceeding a preset time length belongs to an abnormal stop; and performing a traffic anomaly classification on a video frame corresponding to the abnormal stop using a decision tree to obtain a traffic anomaly type, if there is the abnormally stopped vehicle, wherein the decision tree is generated based on features for a traffic anomaly detection.

Methods, systems, and computer-readable storage media for obtaining sensor data captured by one or more sensors that generate data representative of characteristics of a roadway, determining a state of the roadway by processing the sensor data, the state at least partially comprising a first configuration of the roadway, determining, at least partially based on the state, a period of time to deploy a plurality of road barriers on the roadway, and deploying the plurality of road barriers based on the period of time to provide a second configuration comprising at least one dedicated lane.

According the present invention, there is provided a control system (10) that includes an image acquisition unit (11) that acquires an image generated by a camera, a controlled object determination unit (12) that analyzes the image and determines at least one of a vehicle that satisfies a predetermined condition and a vehicle in which a predetermined person is riding included in the image, as a vehicle to be controlled, a control content decision unit (13) that decides a control content for the vehicle to be controlled, and an output unit (14) that outputs a control command including the control content to the vehicle to be controlled.

According the present invention, there is provided a control system (10) that includes an image acquisition unit (11) that acquires an image generated by a camera, a controlled object determination unit (12) that analyzes the image and determines at least one of a vehicle that satisfies a predetermined condition and a vehicle in which a predetermined person is riding included in the image, as a vehicle to be controlled, a control content decision unit (13) that decides a control content for the vehicle to be controlled, and an output unit (14) that outputs a control command including the control content to the vehicle to be controlled.

An apparatus for recording license plates of vehicles travelling on a road having several adjacent lanes comprises a vehicle classification sensor configured to detect a predetermined shape characteristic of a vehicle or group of vehicles. The apparatus further comprises at least one camera mounted at an elevated point beside one of the lanes and having an angle of aperture covering at least one of said lanes, each lane covered by at least one camera. The vehicle classification sensor is configured to, upon detecting the predetermined shape characteristic on a lane, trigger the camera that covers the lane the predetermined shape characteristic is detected on to record an image of a license plate on the back of the vehicle or group's leading vehicle, respectively, for which the predetermined shape characteristic is detected. The triggered camera is of a lane either adjacent to or at least one lane apart from the lane.

An apparatus for recording license plates of vehicles travelling on a road having several adjacent lanes comprises a vehicle classification sensor configured to detect a predetermined shape characteristic of a vehicle or group of vehicles. The apparatus further comprises at least one camera mounted at an elevated point beside one of the lanes and having an angle of aperture covering at least one of said lanes, each lane covered by at least one camera. The vehicle classification sensor is configured to, upon detecting the predetermined shape characteristic on a lane, trigger the camera that covers the lane the predetermined shape characteristic is detected on to record an image of a license plate on the back of the vehicle or group's leading vehicle, respectively, for which the predetermined shape characteristic is detected. The triggered camera is of a lane either adjacent to or at least one lane apart from the lane.

An image processing device 10 includes an image interface 18, a memory 19, and a controller 20. The image interface 18 acquires a captured image. The positions of specific feature points in a world coordinate system and reference positions of the specific feature points are stored in the memory 19. The controller 20 detects the specific feature points in the captured image. In a case where discrepancy between the position in the captured image and the reference position is found with regard to a predetermined percentage or more of the specific feature points, the controller 20 recalculates a calibration parameter.

In some implementations, a system may receive a cable map for a deployed cable. The system may receive vibration data indicating a vibration associated with a first section of the cable. The system may determine a characteristic associated with the first section of the cable based on the vibration. The system may determine a location associated with the characteristic based on the cable map. The system may determine that the first section of the cable is associated with the location based on the location being associated with the characteristic. The system may associate the location and a length of a second section of the cable extending from an initial location to the location. The system may receive an input identifying the length of the second section of the cable and may output the location based on associating the location and the length of the second section of the cable.

A vehicle behavior of a monitored vehicle is learned. A vehicle illumination of the monitored vehicle is detected and monitored. If a light-pattern occurs in the detected vehicle illumination, wherein the light-pattern corresponds to a frequency, intensity and/or color dependent glowing of the vehicle illumination, and further wherein the light-pattern starts with a flashing up of the detected vehicle illumination and ends after a certain time without glowing of the respective part of the detected vehicle illumination, then the method further monitors the light-pattern; monitors a vehicle movement of the monitored vehicle during the occurrence of the light-pattern; and compares the monitored light-pattern with a known light-pattern from a light-pattern data entry stored in an light-pattern database. If the comparison results in the monitored light-pattern being unknown, the method stores the light-pattern and the vehicle movement together as a new light-pattern data entry in the light-pattern database.

The present disclosure provides a vehicle passenger flow statistical method, including: acquiring, by a server, door opening time and door closing time of a vehicle during any one door opening and closing process of door opening and closing processes in a pending statistical period; acquiring, by the server, a first set of images and a second set of images in a carriage of the vehicle according to the door opening time and the door closing time; and acquiring, by the server, the number of on-board people and the number of off-board people during the door opening and closing process according to the first set of images and the second set of images. The present disclosure can improve efficiency and accuracy of passenger flow statistics according to which an operation plan of the vehicle may be optimized timely and effectively.