Mining locomotive pedestrian detection method based on multi-information fusion

Abstract

The invention provides a mining locomotive pedestrian detection method based on multi-information fusion. The method comprises: establishing pedestrian data sets of visible light and infrared light inan underground coal mine; performing CLAHE finite contrast adaptive histogram equalization and optimal notch denoising method processing based on a third-order Butterworth function on the two kinds of images for the underground special environment; adjusting and improving a YOLOv3 target detection network by adopting dense connection and a multi-scale pooling structure; extracting and fusing features of the two types of images; optimizing a loss function by adopting a cross entropy function; model establishment, improving the accuracy, the real-time performance and the stability of the modelin an underground complex environment. Meanwhile, the training precision of the network is further improved by adopting a transfer learning method, and the training time is shortened, so that the pedestrian target is detected by using the target detection model obtained by training, the pedestrian detection result in front of the mining locomotive is output in real time, and the detection speed required by the running of the mining locomotive can be met.

Description

technical field [0001] The invention relates to the technical field of coal mine underground detection, in particular to a method for detecting pedestrians on mining locomotives based on multi-information fusion. Background technique [0002] As the coal mine resource market continues to rise, the underground transportation task is becoming more and more arduous. According to data, due to the influence of factors such as dust and light in the working environment, the rate of safety accidents caused by locomotive transportation in mines accounts for 20% to 30% of the total accidents. During the operation of mine locomotives, transportation accidents may occur due to drivers’ fatigue driving, improper operation, or miners’ illegal operations, which will seriously endanger the life safety of miners and also bring great losses to the production efficiency of the mining industry. [0003] Usually, installing target detection devices on locomotives in coal mines is the main means...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the above problems, the present invention provides a mine locomotive pedestrian detection method based on multi-information fusion, which can solve the problem that existing visible light sensors are easily affected by light and dust in special underground environments, and has strong adaptability and anti-interference ability. And further improve the detection accuracy and real-time

Images