Research On Coal Gangue Identification Based On Deep Learning

With the continuous improvement demands of cleaning and intellectualization of coal gangue separation,the concept of automatic separation of coal and gangue was put forward by domestic researchers.One of the technical difficulties to realize the automatic separation of coal and gangue is to accurately complete the recognition and localization of gangue images.At present,some scholars have carried out extensive research in the aspect of coal gangue target recognition,but the traditional method of coal gangue recognition has some problems such as difficulty in feature extraction,weak generalization ability and unrealized localization.Aiming at the above problems,a method of coal gangue recognition based on deep learning is proposed in this paper,in order to improve the accuracy of coal gangue recognition and localization.In order to collect the image data of coal gangue,the actual production situation of coal gangue separation was analyzed in this paper and the experimental equipment for image data acquisition of gangue was designed by simulating manual separation of gangue;the principle of image collection of coal gangue was put forward,the collection scheme of coal gangue image data set was designed and the pretreatment method of data set was studied.In view of the problem of coal and gangue target recognition,a coal gangue target recognition network based on convolution neural network was constructed by studying the structure and characteristics of convolution neural network in depth learning model,to research the effects of network layer number,convolution kernel size,activation function,pooling method and Dropout on the identification performance of coal and gangue.On the basis of coal gangue target recognition,aiming at the problem of coal and gangue target location,in this paper,the existing methods of generating different candidate regions were analyzed,and a method to generate coal and gangue target candidate regions based on convolution neural network was put forward.The role of sliding window and Anchor Mechanism in the formation Network of Coal gangue Target candidate area was studied to optimize the target candidate areas for coal and gangue and improve the accuracy of location.The experimental results show that the structure of the whole network is reasonable,the of locationThe experimental results show that the structure of the whole network is reasonable,the accuracy and efficiency of identifying location is high.The research results laid a foundation for the industrial application of automatic separation of coal gangue.