Research on Automatic Fault Detection of Coal Mine Underground Conveyor Belt Based on Machine Vision
The large deviation ratio of existing methods for detecting faults in coal mine underground conveyor belts,resulting in low detection accuracy.Therefore,this paper studies the automatic detection of faults in coal mine underground conveyor belts based on machine vision.For coal mine underground conveyor belts,a depth camera is selected to capture high-quality color images,and HDR is used to brighten the images of coal mine underground conveyor belts,restore the details of darker parts,and perform gamma correction on the collected images to enhance the contrast of the images.By applying affine transformation,the surface feature images of the conveyor belt are concatenated frame by frame to reflect the motion relationships between the images.Using the surface feature images of the spliced conveyor belt,analyze the width of the conveyor belt,obtain its offset ratio,and perform offset judgment to complete automatic detection of conveyor belt faults.The experimental results show that the error between the offset ratio of the experimental group and the actual value is small,resulting in better image straight line fitting effect.By statistically analyzing the correct frame rates of video samples A-F,the calculated detection accuracy is above 98%,achieving more accurate detection results and improving the real-time performance of fault detection.
NETL
NSTL
万方数据
