Coal conveying quantity detection of mine belt conveyor based on MT-CNN

In order to realize the informatization and intelligence of coal conveying quantity detection of mine belt conveyor, the coal conveying quantity detection technology of mine belt conveyor based on MT-CNN is proposed. The multi-task convolutional neural network（MT-CNN） is selected to perform multi-core recognition and detection of the detection target, optimize the extraction efficiency of image straight line information and edge information, build a good network hierarchy, optimize the information connection channel, and improve the effect of image recognition analysis and data detection and processing. The contour morphology and load state of the coal conveying quantity are analyzed and calculated by MT-CNN technology, and the relevant data of the coal conveying quantity of the mine belt conveyor are obtained through image sample data training. The experimental results show that the research technology can effectively improve the authenticity of the image recognition of coal conveying quantity, shorten the detection time by 49%, and increase the accuracy of the calculation results to 98%, which proves that the research technology can effectively improve the efficiency and accuracy of the coal conveying quantity detection, and has good application performance and good use effect. Strengthening the coal conveying quantity detection of mine belt conveyor can provide a basis for follow-up research, promote the development of related technologies to a large extent, and realize the development of mine information, intelligence and modernization.