Research on identification of coal and gangue under complex working conditions based on relief feature

ABSTRACTTo improve the identification accuracy of coal gangue under the complex working conditions of pulverized coal and moisture adhesion on the surface, this paper studies the distribution difference of characteristic curves under five working conditions of different proportions of moisture and pulverized coal, and finds that moisture has a greater influence on gradient features, while pulverized coal has a greater influence on gray texture features. Then this paper studies the distribution of coarse texture and fine texture in coal and gangue, and finds that there are many fine textures and few coarse textures on the surface of gangue, while coal shows the opposite law. In addition, the ratio of pixels with large gray value to pixels with small gray value can reduce the occurrence of noncompliance with the above laws. Finally, through the combination of coarse texture, fine texture, and gray value difference, the relief feature is put forward. Compared with different features, the average feature overlap ratio proposed in this paper is lower, at 23.83%. Compared with different methods, this method has a higher identification rate under different working conditions, and the average identification accuracy rate is 89.59%, which is at least 13.93% higher than other methods.KEYWORDS: Coal and ganguecomplex working conditionscoal gangue identificationrelief featurecoarse and fine textures AcknowledgementsThis research was funded by the Institute of Energy, Hefei Comprehensive National Science Center(Grant No. GXXT-2022-16), Scientific Research Foundation for High-level Talents of Anhui University of Science and Technology(Grant No. 2023yjrc56), National Natural Science Foundation of China Project (Grant No. 52274152), Anhui Province University Outstanding Youth Research Project(Grant No. 2022AH020056).Disclosure statementThe authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper.CRediT authorship contribution statementQisheng Luo: Methodology, Software, Writing an original draft. Shuang Wang: Software, Writing-review & editing, Funding acquisition. Yongcun Guo and DeYong Li: Conceptualization, Writing-review & editing, Funding acquisition. Lei He and Gang Cheng: Conceptualization, Writing-review & editing.Additional informationFundingThis work was supported by the The Institute of Energy, Hefei Comprehensive National Science Center [GXXT-2022-16]; Scientific Research Foundation for High-level Talents of Anhui University of Science and Technology [2023yjrc56]; National Natural Science Foundation of China Project [52274152]; Anhui Province University Outstanding Youth Research Project [2022AH020056].