Research on coal gangue recognition method based on infrared thermal imaging

Coal and gangue sorting methods based on heavy-medium coal selection technology, jigging technology, flotation technology, dry coal selection technology and γ-ray detection method have high investment costs, low sorting efficiency and serious environmental pollution. The accuracy of the coal gangue sorting method based on CCD cameras is not high, and the X-ray based coal gangue sorting technology can harm the health of personnel. Infrared thermal imaging technology has the advantage of being unaffected by light and dust, and will not cause harm to the human body. A coal gangue recognition method based on infrared thermal imaging has been proposed. Firstly, coal and gangue pass through the heating area under the conveyor belt, and the temperature of the center point of coal and gangue is monitred through an infrared thermal imager to obtain the temperature of the heated coal and gangue. The infrared thermal imager is used to capture the uniformly heated coal and gangue in the heating area, obtaining infrared grayscale and color images of the coal and gangue. Secondly, Gaussian filtering is used to preprocess and extract features from the infrared grayscale images and infrared color images of coal and gangue. The grayscale mean of the infrared grayscale image, the grayscale value feature corresponding to the maximum frequency, and the G-channel first-order moment and G-channel second-order moment features of the infrared color image are used as sorting features. The above four features are used as inputs for the classification model. Finally, support vector machine (SVM) is used for classification and recognition to achieve the goal of recognizing coal and gangue. The experimental results show that the coal gangue recognition method based on infrared thermal imaging has achieved an accuracy rate of over 98% for the sorting of bituminous coal, anthracite, and lignite, and has a good classification effect.