State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mines, Anhui University of Science and Technology
School of Mechanical and Electrical Engineering, Anhui University of Science and Technology

基于重介选煤、跳汰选煤、浮选、干法选煤、γ射线检测法的煤矸分选方法投资成本高、分选效率低、环境污染严重，基于CCD相机的煤矸分选方法准确率不高，基于X射线的煤矸分选技术会危害工作人员的健康。红外热成像技术不受光照、粉尘影响，且不会对人体造成伤害。提出了一种基于红外热成像的煤矸识别方法。首先，煤和矸石在传送带的输送下经过加热区域，红外热成像仪监测经均匀加热后的煤和矸石中心点的温度，得到煤和矸石加热后的温度并对经加热区域均匀加热后的煤和矸石进行拍摄，得到煤和矸石的红外灰度图像和红外彩色图像。然后，选用高斯滤波对煤和矸石的红外灰度图像、红外彩色图像进行预处理并提取特征，将红外灰度图像的灰度均值、最大频数对应的灰度值特征和红外彩色图像的G通道一阶矩、G通道二阶矩特征作为分选特征，将上述4个特征作为分类模型的输入。最后，采用支持向量机（SVM）进行分类识别，从而达到识别煤和矸石的目的。实验结果表明：基于红外热成像的煤矸识别方法对烟煤、无烟煤、褐煤的分选准确率均达到了98%以上，有良好的分类效果。

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.

安徽省高校协同创新项目（GXXT-2021-076）。

程刚，潘择烨，魏溢凡，等. 基于红外热成像的煤矸识别方法研究[J]. 工矿自动化，2024，50（4）：69-77.

CHENG Gang, PAN Zeye, WEI Yifan, et al. Research on coal gangue recognition method based on infrared thermal imaging[J]. Journal of Mine Automation，2024，50（4）：69-77.