Detection of Underground Dangerous Area Based on Improving YOLOV8

Electronics, Journal Year: 2024, Volume and Issue: 13(3), P. 623 - 623

Published: Feb. 2, 2024

In order to improve the safety needs of personnel in dark environment under well, this article adopts improved YOLOV8 algorithm combined with ray method determine whether underground are entering dangerous areas and provide early warning. First all, introduces coordinate attention mechanism on basis target detection so that model pays location information area as accuracy obstruction small areas. addition, Soft-Non-Maximum Suppression (SNMS) module is introduced further accuracy. The then be deployed applied a variety angles different scenic cameras. experimental results show proposed obtains 99.5% identification frame speed 45 Frames Per Second (FPS) self-built dataset. Compared model, it has higher can effectively cope changes interference factors environment. Further, meets requirements for real-time testing

Language: Английский

---

Deep Reinforcement Learning for automated scheduling of mining earthwork equipment with spatio-temporal safety constraints

Frontiers of Engineering Management, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 24, 2025

Language: Английский

---

Enhancing Object Detection in Underground Mines: UCM-Net and Self-Supervised Pre-Training

Sensors, Journal Year: 2025, Volume and Issue: 25(7), P. 2103 - 2103

Published: March 27, 2025

Accurate real-time monitoring of underground conditions in coal mines is crucial for effective production management. However, limited computational resources and complex environmental mine shafts significantly impact the recognition capabilities detection models. This study utilizes a comprehensive dataset containing 117,887 images from five common mining tasks: personnel detection, large lump identification, conveyor chain monitoring, miner behavior recognition, hydraulic support shield inspection. We propose ESFENet backbone network, incorporating Global Response Normalization (GRN) module to enhance feature capture stability while employing depthwise separable convolutions HGRNBlock modules reduce parameter volume complexity. Building upon this foundation, we UCM-Net, model based on YOLO architecture. Furthermore, self-supervised pre-training method introduced generate mine-specific pre-trained weights, providing with more semantic features. utilizing combined neck portions as encoder an image-masking structure strengthen acquisition improve performance small models learning. Experimental results demonstrate that UCM-Net outperforms both baseline state-of-the-art YOLOv12 terms accuracy efficiency across datasets. The proposed architecture achieves 21.5% reduction 14.8% load decrease compared showing notable improvements 1.3% (mAP50:95) 0.8% (mAP50) recognition. framework effectively enhances training efficiency, enabling attain average mAP50 94.4% all research outcomes can provide key technical safety offer valuable technological insights public sector.

Language: Английский

---

Using Generative Pre-Trained Transformer-4 (GPT-4), ffmpeg, and Microsoft Azure to Aid in Creating a Text-to-Video Generation Tool to Improve Safety Shares and Incident Descriptions in the Mining Industry

Mining Metallurgy & Exploration, Journal Year: 2025, Volume and Issue: unknown

Published: March 27, 2025

Language: Английский

---

Revolutionizing Open-Pit Mining Fleet Management: Integrating Computer Vision and Multi-Objective Optimization for Real-Time Truck Dispatching

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(9), P. 4603 - 4603

Published: April 22, 2025

The implementation of fleet management software in mining operations poses challenges, including high initial costs and the need for skilled personnel. Additionally, integrating new with existing systems can be complex, requiring significant time resources. This study aims to mitigate these challenges by leveraging advanced technologies reduce minimize reliance on highly trained employees. Through integration computer vision multi-objective optimization, it seeks enhance operational efficiency optimize open-pit mining. objective is truck-to-excavator assignments, thereby reducing excavator idle deviations from production targets. A YOLO v8 model, six hours mine video footage, identifies vehicles at excavators dump sites real-time monitoring. Extracted data—including truck assignments ready times—is incorporated into a binary integer programming model that waiting times discrepancies target assignments. epsilon-constraint method generates Pareto frontier, illustrating trade-offs between objectives. Integrating image analysis optimization significantly improves efficiency, enabling adaptive truck-excavator allocation. highlights potential techniques mining, leading more cost-effective data-driven decision-making.

Language: Английский

---

Vulnerability and Risk Management to Ensure the Occupational Safety of Underground Mines

Eng—Advances in Engineering, Journal Year: 2025, Volume and Issue: 6(5), P. 88 - 88

Published: April 25, 2025

Ensuring occupational safety in underground mines is a fundamental priority due to the major risks associated with this unfriendly work environment. This involves employing set of technical, organizational, and educational measures reduce hazards for workers minimize accidents diseases electrical mechanical causes. Old precarious coal extraction methods, conjunction obsolete infrastructure installations, lead high accident risk, endangering lives when at work. Precarious working conditions materials alongside carelessness decision makers make mine-based cause professional illnesses. In paper, authors identify, estimate, prioritize, evaluate vulnerabilities within discuss actions resources necessary mitigate, stop, and/or eliminate these vulnerabilities, as well mitigation strategy stopping eliminating them achieve increased safety.

Language: Английский

---

Application of Artificial Intelligence in Virtual Reality

Algorithms for intelligent systems, Journal Year: 2024, Volume and Issue: unknown, P. 67 - 85

Published: Jan. 1, 2024

Language: Английский

---

Innovative predictive maintenance for mining grinding mills: from LSTM-based vibration forecasting to pixel-based MFCC image and CNN

The International Journal of Advanced Manufacturing Technology, Journal Year: 2024, Volume and Issue: 135(3-4), P. 1271 - 1289

Published: Oct. 8, 2024

Language: Английский

---

Applications of Machine Vision in Coal Mine Fully Mechanized Tunneling Faces: A Review

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 102871 - 102898

Published: Jan. 1, 2023

The realization of intelligent mining is the only method for realizing high-quality development in coal industry. As forefront working link mine production, achieving automatic roadway tunneling control key to improving production efficiency, enhancing intelligence, and reducing accident rates at fully mechanized faces. Among various detection techniques, machine vision technology stands out with advantages non-contact measurement, rich information acquisition, high accuracy. equipment groups based on has become a research hotspot intelligence process mines. This study first introduces technologies visual system, including camera calibration, image preprocessing, feature extraction, matching, target segmentation recognition, 3D reconstruction. It then elaborates principles, workflows, limitations, precautions, status systems practical application scenarios faces, such as equipments, anchoring systems, transportation safety auxiliary which significantly improve efficiency. Finally, considering challenging work conditions strong interference mines, successful adaptation excavation sites relies addressing technical challenges related poor environment adaptability, limited imaging field view, low level. Furthermore, according existing results current status, this paper forecasts that need be developed future vision, multi-sensor fusion, group collaborative control, digital twin-driven remote monitoring.

Language: Английский

---

Design of coal mine drilling detection model combining improved YOLOv5 and Gaussian filtering

Energy Informatics, Journal Year: 2024, Volume and Issue: 7(1)

Published: Sept. 30, 2024

Language: Английский

---