Automatic Identification Of Coal And Gangue Based On Machine Vision

Coal is a very important strategic energy and is one of the indispensable fossil fuels for the development of the country.In the process of coal mining,gangue as impurity with coal as is mined.The existence of gangue reduces the quality of coal and affects the utilization efficiency of coal.Therefore,the separation of coal and gangue is an indispensable part in the process of coal mining.With the rapid development of machine vision technology,aiming at the disadvantages of traditional coal selection,a new method of coal and gangue sorting based on image processing was studied in this paper.The actual coal mining environment will have a large of coal dust and noise to affect the quality of the image,and the collected coal and gangue should be pretreated.After comparison,the median filter was selected to enhance the image,and the coal and gangue in the image ware separated from the background by the.threshold segmentation method.Then two algorithms of fusion LBP and GLCM were proposed to extract the texture features of coal and gangue images.By selecting the LBP operator with rotation invariance,the original image was transformed into a local binary image,and the grayscale symbiosis matrix of the image was reproduced.Choose the angular second moment,contrast,correlation and entropy of the four.texture characteristics as indicators of extraction,the mean and normalized after processing,the use made of the radial basis kernel function of support vector machine for training and recognition.This algorithm was used to select a certain number of experimental samples for analysis,and the experimental results verify the effectiveness of the algorithm.Based on the recognition algorithm,this paper designed a set of automatic sorting system for coal and gangue.The system was mainly composed of hardware and software,the hardware included four parts:queuing device,image acquisition mechanism,image processing recognition mechanism and control actuator;the software system was mainly developed based on OpenCV visual library and.Net 4.0 platform.The system was designed and developed on the basis of machine vision,which not only can effectively detect the extracted coal in real time,but also can realize non-contact measurement.Compared with the traditional method of coal preparation,it takes up small space,simple operation,low cost,easy maintenance,and saves a lot of resources and protects the environment.The design scheme of this paper provided a valuable reference for the separation of coal and gangue in actual coal mining.