Abstract:A large number of coal mines are widely distributed over China. Bulk coal gangue deposits seriously affect the mining area environment, and some mishandling of coal gangue may cause spontaneous combustion and explosion, which poses a direct threat to mine safety. The comprehensive utilization of coal gangue can effectively alleviate this problem, and it is of significance to the ecological safety and sustainable development of mine. Depending on the burning state, coal gangue is divided into two types - burned and unburned gangue, whose hidden dangers of security and harm to the environment are different, as well as ways of comprehensive utilization. Therefore, it is very important to do the classification, recognition and monitoring of the coal gangue. The current monitoring methods are mainly the field investigation with low efficiency and high cost, almost impossible for meeting the actual demand of coal gangue monitoring. Tiefa mine in Liaoning Province was chosen as the study area. Firstly, a total of 106 typical coal gangue samples were collected from waste dump in mining areas. Then, SVC HR1024 spectrometer was used to test the visible and near infrared spectrum of samples, and a differential spectral index NDGI was constructed to identify the burned and unburned gangue based on the difference of spectral characteristics of the burned and unburned gangue. Finally, the laboratory spectral data and the corresponding satellite remote sensing images were utilized for verifying the index. The random forest classification method was used as a contrast to the results of the laboratory spectrum treatment. The results showed that the slope of the spectral curves of burned gangue samples was higher ranging from 350 to 750 nm, and the reflectance within range of 550 similar to 630 nm increased sharply, while the slope of the unburned gangue in the whole visible bands of spectrum was lower. The threshold of the NDGI index was set as 0. 25 to distinguish the burned and unburned gangue. The laboratory spectral data showed that the classification accuracy of the NDGI index is up to 99. 1%, higher than that of 95. 2% of the random forest classification method. The Field results showed burned and unburned areas of waste dump were distinguished and classified in Landsat8 OLI images based on the NDGI index, and the burned and unburned coal gangue areas were in good agreement with the Google Earth on the morphology and size. The overall results showed that the index can effectively distinguish the combustion states of gangue. In addition, burned and unburned gangue samples were taken for mineral identification respectively. By comparing the changes of mineral species before and after combustion, the cause of spectral difference was analyzed between the burned and unburned gangue. The results showed the oxidation from Fe(2+)to Fe (3+) of gangue in the process of combustion. A large increase in Fe3+ caused the formation of an obvious spectral valley characteristic at 550 nm band, and a highly reflectance appeared at 750 nm band due to the glass quality generated during combustion. The above conditions cause differences in NDGI index between the burned and unburned gangue. In this paper, the results provide a fast, efficient and accurate model and method for burned and unburned gangue distinguishment in coal mine.

Dual-energy X-ray transmission identification method for coal and gangue with different thicknesses and densities

Dual-energy X-ray transmission identification method of multi-thickness coal and gangue based on SVM distance transformation

Identification of coal and gangue based on R value method for dual-energy X-ray of Geant4 simulation

Study of raw coal identification method by dual-energy X-ray and dual-view visible light imaging

Identification of Coal and Gangue Using Visible-Near Infrared (Vis-NIR) Hyperspectral Imaging

A novel multi-dimensional coal and gangue X-ray sorting algorithm based on CdZnTe photon counting detectors

A Classification Method Based on the Combination of Visible,Near-Infrared and Thermal Infrared Spectrum for Coal and Gangue Distinguishment

Coal and Gangue Classification Based on Laser-Induced Breakdown Spectroscopy and Deep Learning

Coal gangue recognition based on spectral imaging combined with XGBoost

Coal Gangue Classification Based on the Feature Extraction of the Volume Visual Perception ExM-SVM

Coal and gangue classification in actual environment of mines based on deep learning

Research on visual features and identification of coal and gangue in simulated underground mine environments

Multiple factors influence coal and gangue image recognition method and experimental research based on deep learning

Automatic Identification of Coal and Rock/Gangue Based on DenseNet and Gaussian Process

Image feature extraction and recognition model construction of coal and gangue based on image processing technology

A coal and gangue detection method for low light and dusty environments

A Classification Method Based on the Visible Spectrum for Burned and Unburned Gangue Distinguishment

Precise detection of coal and gangue based on natural γ-ray

Coal and Gangue Active Identification Method Using Microwave Irradiation-Infrared Detection

Detection of coal content in gangue via image analysis and particle swarm optimization–support vector machine

Feature Extraction with Multi-fractal Spectrum for Coal and Gangue Recognition Based on Texture Energy Field