Vehicle queuing length detection method based on motion detection

Abstract

The invention discloses a vehicle queuing length detection method based on motion detection. The vehicle queuing length detection method comprises a first step of calibrating lane information and setting a vehicle static threshold value; a second step of reading an image sequence, and intercepting a vehicle detection area G in a current frame; a third step of tracking a vehicle, and updating the position of the tracked vehicle at the current frame; a fourth step of detecting newly appeared vehicles in the f; a fifth step of finding vehicles in the static state in each lane; a sixth step ofcalculating the queuing length of each lane, and converting pixel coordinates with the farthest distances of each lane in the QL into vehicle queuing length distance according to the actual distance calibrated by the lane lines; and a seventh step of calculating the traffic flow of each lane. The queuing accuracy of vehicles is not affected by excessively large distance between the queuing vehicles, and the accurate queuing length is calculated through the farthest position of the static vehicle in each lane, and the vehicle flow of each lane can be obtained at the same time.

Description

technical field [0001] The invention relates to the field of intelligent transportation, in particular to a method for detecting vehicle queuing length based on motion detection. Background technique [0002] The collection of traffic information such as vehicle queue length and traffic flow based on video is the application of image processing technology in the field of intelligent transportation. This technology is widely used for its advantages of convenient installation and maintenance, strong intuitiveness, expansibility and large room for improvement. It is the development direction of traffic flow information detection technology in the future. Its research and application have important theoretical significance and use value. [0003] In the process of calculating the vehicle queuing length, the current commonly used detection methods often calculate the vehicle queuing length by analyzing static images. Among them, the closest technical solution to the present inve...

AI Technical Summary

Problems solved by technology

However, when the vehicle queuing distance is too large, the method cannot adapt to the diversity of queuing forms; ):100-109) proposed a comprehensive detection method based on edge information and local texture features. This method first uses the Canny algorithm to extract vehicle edge information, and uses the improved LBP method to extract vehicle texture features. Carry out segmentation to obtain the stationary vehicle queue and perform horizontal projection to calculate the queue length, but this method is prone to interference from road shadows and lane lines when extracting vehicles, and the obtained moving vehicles and stationary vehicles are incomplete, resulting in low accuracy of queue length , and this method cannot obtain the traffic flow of each lane at the same time

Images