Deep learning and image processing fused traffic signal lamp fault detection method

Abstract

The invention discloses a deep learning and image processing fused traffic signal lamp fault detection method. The method comprises the following steps: S1, obtaining an electronic police video streamunder various weather conditions, and decoding the electronic police video stream to obtain a traffic signal lamp image, position information of a traffic signal lamp group in the image, and type information and position information of each signal lamp in the traffic signal lamp group; S2, configuring traffic signal lamp group information and information of each signal lamp in the lamp group; andS3, establishing a traffic signal lamp detection model. According to the invention, the problems of jittering and position deviation of a small-range camera are effectively solved, and the alarm accuracy is improved; and a deep learning technology based on a convolutional neural network is adopted to realize position and state identification of the signal lamp, and an image processing algorithm based on a signal lamp period is adopted to further improve the identification accuracy of the signal lamp.

Description

technical field [0001] The invention relates to the technical field of traffic signal lamp status monitoring, in particular to a traffic signal lamp fault detection method integrated with deep learning and image processing. Background technique [0002] Traffic lights are a category of traffic safety products that can be used to strengthen road traffic management and improve road use efficiency. The normal operation of traffic lights is the basis for the normal operation of the city, but at present, a large number of existing traffic lights are non-intelligent traffic lights without fault self-diagnosis capabilities. The fault detection of such traffic lights is mainly through the maintenance of the traffic police on duty, and the patrol inspection of traffic facilities units. and public alarms. [0003] The above signal lamp fault detection method has problems such as heavy workload of maintenance personnel, untimely fault detection, and low efficiency. In view of the act...

AI Technical Summary

Problems solved by technology

[0006] The present invention mainly solves the problems of weak detection model generalization ability, low model recognition accuracy, and missed detection in the detection process in the existing video detection method; it provides a traffic signal light fault detection method integrated with deep learning and image processing, which reduces Improve the false alarm rate based on video detection signal lamp failure, improve the accuracy of the alarm

Images