Prohibited parking detection method based on intersection-parallel ratio

Abstract

The invention discloses a prohibited parking detection method based on an intersection-parallel ratio, which comprises the following steps: 1, taking one frame of picture every N frames, detecting a vehicle and other targets through a deep learning neural network detection algorithm and a supervised learning mechanism, and extracting the coordinates of the left lower corner and the right upper corner of a rectangular frame of a vehicle target; 2, calculating the intersection-parallel ratios of all target rectangular frames in the current frame of picture and the previous frame of picture, andsorting the intersection-parallel ratios; and 3, when the detected intersection-parallel ratio values are greater than a threshold value and are greater than the threshold value for multiple consecutive times, returning the coordinates of the target object and the rectangular frame, and judging that the vehicle is parked. According to the method, an abnormal parking event can be accurately judged,and a relatively accurate detection method is provided for detecting abnormal parking events through a video on the urban expressway.

Description

technical field [0001] The invention belongs to the fields of artificial intelligence and intelligent traffic, and proposes a detection method for prohibited parking based on cross-combination ratio. Background technique [0002] Traditional illegal parking detects the foreground target through the difference between the video image and the background image, and then judges the stationary target through pixel-level time series analysis, and then identifies the prohibited vehicle through feature extraction. [0003] With the maturity of the deep neural network technology, it has also appeared that the vehicle is detected first through the neural network, and then through the change of the center position of the vehicle target between consecutive frames, if the position changes little or no change, the vehicle can be determined to stop. However, when this solution is actually applied, the frame of the detected vehicle will vibrate, causing the center of the vehicle to vibrate,...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problems of jittering phenomenon and high missed detection rate, the present invention proposes a detection method for prohibited parking based on a supervised learning neural network to detect vehicles and based on the cross-combination ratio

Images