Mine track obstacle detection method based on information fusion

Abstract

As an important part of the coal mine transportation system, coal mine underground rail transportation undertakes the core transportation task of coal mine underground. Its safe and efficient operation is directly related to the efficiency of coal mine production and transportation. In view of this, this paper proposes a mine track environmental obstacle detection system which integrates camera and Lidar information to realize real-time automatic detection of obstacles in front of the underground track mine cart. The specific research results are as follows: first of all, a point cloud clustering algorithm for the mine environment is designed to extract the obstacle information, and then the YOLOv5 algorithm is used to identify the obstacle information in the image. Finally obstacle information from image and point cloud are fused at the decision-level. The obstacle detection method proposed in this paper can be successfully identified to meet the requirements of the obstacle object detection function of the rail vehicle.

The Future of Mine Safety: A Comprehensive Review of Anti-Collision Systems Based on Computer Vision in Underground Mines

“…In response to the need for real-time detection of obstacles in front of underground track mine carts, a new system [32] was proposed that utilizes both camera and LIDAR information. The system employs a custom point cloud clustering algorithm specifically designed for the challenging mine environment to extract obstacle information and then leverages the YOLOv5 algorithm to identify obstacles in the resulting images.…”

“…Yolo (v2,v3,v4 and v5) [21,22,[29][30][31][32] RGB Images [21,22,[30][31][32], Thermal Images [22,29] Pedestrian detection [21,22,[29][30][31][32], electric locomotives and stones falling [30] HOG [22] RGB Images [22], Thermal Images [22] Pedestrian detection [22] SVM [90,100] RGB Images [90,100], Thermal Images [100] Enhancing underground visual place [90], pedestrian segmentation [100] Image Segmentation and Thresholding [78,100,101] RGB Images [78], Thermal images [78,100,101] Overhead boulders detection [78], pedestrian detection [100,101] Navigation and mapping [23][24][25][26][27][28] RGB Image…”

“…Perhaps one of the most promising applications of AI in mining is in the development of anti-collision systems based on computer vision. These systems use advanced sensors, such as cameras and LIDAR [19,20], along with sophisticated algorithms and deep learning techniques to detect and track the movement of people and objects in underground mines [21][22][23][24][25][26][27][28][29][30][31][32]. This technology can help prevent accidents and injuries by providing real-time alerts and warning signals to operators and workers [22].…”

Detection of Underground Dangerous Area Based on Improving YOLOV8

“…By locating the track and extending it a certain distance outside the track, unsafe areas for electric locomotives to travel can be found. In response to the need for the real-time detection of obstacles in front of underground rail mine cars, Biao et al [19] proposed a new system that simultaneously utilizes camera and lidar information. The system uses a custom point cloud clustering algorithm designed for challenging mine environments to extract obstacle information and then uses the YOLOv5 algorithm to identify obstacles in the generated images.…”