Image segmentation has been increasingly used to identify the particle size of crushed ore. How to accurately identify the ore particles in complex a environment is particularly important.

In the mining process of underground metal mines, the misjudgment of rock types by on-site technicians will have a serious negative impact on the stability evaluation of rock mass and the formulation of support schemes, which will result in the loss of economic benefits and potential safety hazards of mining enterprises. In order to realize the precise and intelligent identification of rock types, the image data of peridotite, basalt, marble, gneiss, conglomerate, limestone, granite, magnetite quartzite are amplified. Under the target detection framework of Faster R-CNN deep learning, the extraction network based on simplified VGG16 is used to extract and learn features of rock images, and finally the rock type identification system is successfully trained. The experimental verification shows that the system is correct for single-type rock image recognition and the accuracy is more than 96%. In order to realize accurate and intelligent identification of the surrounding rock surface under complex lithological conditions, the multi-type rocks hybrid images are also identified. The results show that the recognition effect is great and the accuracy rate is over 80%. Therefore, this system can accurately identify rock types with similar image features, which proves that the model has strong robustness and generalization ability. It has broad application prospects in rock mass stability evaluation and rock classification in underground mining.

An in-depth understanding of the gravity flow behaviour of broken ore is beneficial to improving the ore recovery in sublevel caving mines. In this study, granular dolomites were used as an analogue to experimentally explore the effect of the physical properties of fragments on the gravity flow behaviour of broken ore. Five groups of granular dolomites were prepared first. Then, the particle size, particle size distribution, and particle morphology (e.g., sphericity, circularity, and fractal dimension) in each group were measured following the standard particle morphology test. After that, the angle of repose and the isolated extraction zone (IEZ) of each group were determined by performing funnel and isolated draw experiments, respectively. The experimental results highlighted the insubstantial effect of particle size and the significant effect of particle morphology on the angle of repose and isolated extraction zone of granular dolomite. It was found that a wider particle size distribution yielded an isolated extraction zone with a nonuniform shape and a smaller mass drawn because of the percolation of finer particles and the jamming of larger particles. Further, a linear negative correlation was revealed between the repose angle and the IEZ mass. Establishing the relationship between the repose angle and the IEZ shape would be of great significance. In this regard, the findings of this study can aid a preliminary analysis.

Before achieving yielding support with artificial pillars, it is significant to evaluate their active support and passive bearing performances on the stope roof. This paper focuses on three aspects of research using 3DEC numerical simulation, which are support patterns of artificial pillars, magnitude of support stresses, and the magnitude of prestresses of the load‐increasing yielding support pattern. Simulation results show that the superior sequence of supporting effect is load‐increasing yielding support, load‐shedding yielding support, and constant and nonyielding support under the same initial support stress. When the magnitude of support stress or the magnitude of prestress is larger with load‐increasing yielding support, the supporting effect is superior and the load‐increasing yielding support with a lower magnitude of support stress is superior to some other support patterns with higher magnitude of support stresses. The active support can improve the support effect compared with no prestress, and under the same final support stress, the support effect is superior when the roof stress releases more in the early supporting stage regardless of the prestress.

The safety of blasting site in open-pit mine can be greatly improved by risk early warning. Therefore, an intelligent real-time risk early warning method of blasting site in open-pit mine based on deep learning was proposed in this research. The mobile wireless webcams, H.264 video compression algorithm, and real-time transport protocol were applied to achieve real-time video acquisition and transmission of blasting site in open-pit mine. A single-stage deep neural network DG-YOLOv3 was proposed in this research. DG-YOLOv3 is an improvement of Gaussian YOLOv3, among which Darknet41 is used to improve the model’s detection speed and detection accuracy of small targets. To further improve the performance (i.e., speed and accuracy) of risk early warning, the surveillance videos were first split into pictures by frame. Then, the pictures were processed by weighted average grayscale and contrast limited adaptive histogram equalization. Experiments show that the mean average precision of DG-YOLOv3 proposed in this paper reaches 87.45 and the detection speed reaches 56.82 frames per second, which has better accuracy and speed compared with other algorithms. In addition, DG-YOLOv3 has good robustness in complex scenarios. Based on the detection results, the intelligent real-time risk early warning of the blasting site in open-pit mine is achieved finally.