Urban road signal lamp pole inclination fault detection method

Abstract

The invention discloses an urban road signal lamp pole inclination fault detection method. A YOLOv3 detection framework is adopted, input of an original image and output of the position and the category of an object are completed, and a YOLOv3 detection model is obtained by training a manually marked image. The model is used for carrying out target detection on a signal lamp post and a pedestrian crossing line, gray scale transformation is carried out on detected target object areas of the signal lamp post and the pedestrian crossing line to obtain a gray scale image, canny edge detection is carried out on the gray scale image, and area communication is carried out through morphological processing; a Hough transform straight line of the pedestrian crossing line and the signal lamp post target object is extracted, polar coordinate angle calculation is performed on the straight line to obtain a straight line included angle, whether the included angle of the pedestrian crossing line and the signal lamp post exceeds a reasonable threshold range or not is judeged, and whether an inclination phenomenon exists or not is judged according to the included angle. The inclination condition of the signal lamp pole can be monitored in real time, an alarm is given in time after the monitored inclination angle exceeds a threshold value, and the advantages of low cost and simple operation are achieved.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, relates to the application of video monitoring and image processing technology, and in particular relates to a detection method for an inclination fault of an urban road signal light pole. Background technique [0002] In order to facilitate the passage of people and vehicles, signal lights are installed at most crossings of urban roads and pedestrian crossing lines are set, and signal lights are erected on light poles. Some intersections have a large space span. In order to facilitate the drivers to observe the status of the signal lights, the light poles need to span multiple lanes at this time. Affected by external forces (strong wind, large vehicles, road construction, etc.), the light pole may tilt or even fall, so it is necessary to monitor whether the light pole is tilted. [0003] Currently, there are three main monitoring methods: [0004] (1) Manual inspection, that...

AI Technical Summary

Problems solved by technology

[0006] 1. The construction is complicated, and physical devices need to be installed on the signal light pole;

[0007] 2. The detection effect is poor, and the inclination sensor is prone to offset on the light pole, which affects the detection effect;

[0008] 3. The detection range is narrow, and when the light pole is long, it can only detect the vicinity of the installation position, and cannot detect the long-distance position;

[0010] 1. The anti-interference performance is poor. When the image acquisition equipment shakes or the image acquisition equipment deviates, it is impossible to accurately locate the position of the signal lamp panel, which affects the detection;

[0011] 2. The operation is complicated, and the detection needs to calibrate and record the normal position value first, and then compare the real-time detection value with the calibrated normal value;

[0012] 3. Poor adaptability, the inclination detection method of signal lamp horizontal bar and vertical bar is not universal

Images