Search Results (329)

The rapid development of Convolutional Neural Networks (CNNs) has markedly boosted the performance of object detection in remote sensing. Nevertheless, tiny objects typically account for an extremely small fraction of the total area in remote sensing images, rendering existing IoU-based or area-based evaluation metrics highly sensitive to minor pixel deviations. Meanwhile, classic detection models face inherent bottlenecks in efficiently mining discriminative features for tiny objects, leaving the task of tiny object detection in remote sensing images as an ongoing challenge in this field. To alleviate these issues, this paper proposes a tiny object detection method based on Normalized Gaussian Label Assignment and Multi-scale Hybrid Attention. Firstly, 2D Gaussian modeling is performed on the feature receptive field and the actual bounding box, using Normalized Bhattacharyya Distance for precise similarity measurement. Furthermore, a candidate sample quality ranking mechanism is constructed to select high-quality positive samples. Finally, a Multi-scale Hybrid Attention module is designed to enhance the discriminative feature extraction of tiny objects. The proposed method achieves 25.7% and 27.9% AP on the AI-TOD-v2 and VisDrone2019 datasets, respectively, significantly improving the detection capability of tiny objects in complex remote sensing scenarios. Full article

Accurate characterization of ore grindability is essential for optimizing mill throughput, reducing energy consumption, and predicting mill performance under varying ore conditions. However, the standard Bond work index (BWI) test remains time-consuming, costly, and requires a large amount of sample. This study evaluates the effectiveness of several rapid, low-cost alternatives, Leeb rebound hardness (LRH), Cerchar abrasivity Index (CAI), portable X-ray fluorescence (pXRF), and hyperspectral imaging (HSI), as proxies for grindability in gold-bearing ores. Sixty-two hand-size rock samples collected from two adjacent Canadian open-pit mines were analyzed using these techniques and subsequently grouped into ten ore groups for BWI testing. LRH and CAI effectively differentiated moderate (<15 kWh/t) from hard (>15 kWh/t) grindability classes, while geochemical features and HSI-based mineralogical attributes also showed strong predictive capability. HSI, in particular, provided non-destructive, spatially continuous data that are advantageous for complex geology and large-scale operational deployment. A conceptual workflow integrating HSI with complementary field measurements is proposed to support comminution planning and optimization, enabling more responsive and timely decision-making. While BWI testing remains necessary for circuit design, the results highlight the value of combining rapid proxy measurements with advanced analytics to enhance geometallurgical modelling, reduce operational risk, and improve overall mine-to-mill performance. Full article

The current method of manually handling or using open-loop automation to deal with abnormal coal lumps on the scraper conveyor is inefficient due to constraints, such as safety concerns and equipment wear. To address inefficiencies in the handling of abnormal coal blocks on scraper conveyors, a reinforcement-learning-based method is proposed. Aiming to address the issue that experimenting on abnormal coal handling by scraper conveyors is expensive, this paper designs a variational Auto-Encoder model with the U-MLP network as its core to simulate the processing environment. In addition, given the sparse characteristics of coal block point cloud data, a deep reinforcement learning model based on the LKDG model is designed to control the crushing equipment when dealing with abnormal coal blocks. Through the point cloud data, images, and other information collected by the fully mechanized mining laboratory before and after abnormal processing of coal blocks, we built a simulation environment for abnormal coal blocks, and trained the LKDG model in the simulation environment. To validate the proposed model, we compared LKDG with baseline models in simulation experiments. The results demonstrate that this method can effectively enhance the efficiency of abnormal coal lump processing without human intervention: LKDG achieved a 10.92% higher average reward compared to existing approaches. In terms of engineering applicability, the trained LKDG delivered excellent performance in laboratory tests conducted in a fully mechanized mining environment, increasing the effective crushing count by 67.11% over conventional automated processing methods. Full article

Accurate detection of oil and gas (O&G) facilities in high-resolution remote sensing imagery is critical for infrastructure surveillance and sustainable resource management, yet conventional detectors struggle with severe class imbalance, extreme scale variation, and arbitrary orientation. In this work, we propose OGF Oriented R-CNN (Oil and Gas Facility Detection Oriented Region-based Convolutional Neural Network), an enhanced oriented detection model derived from Oriented R-CNN that integrates three improvements: (1) O&G Loss Function, (2) Class-Aware Hard Example Mining (CAHEM) module, and (3) Feature Pyramid Network with Feature Enhancement Attention (FPNFEA). Working in synergy, they resolve the coupled challenges more effectively than any standalone fix and do so without relying on rigid one-to-one matching between modules and individual issues. Evaluated on the O&G facility dataset comprising 3039 high-resolution images annotated with rotated bounding boxes across three classes (well sites: 3006, industrial and mining lands: 692, drilling: 244), OGF Oriented R-CNN achieves a mean average precision (mAP) of 82.9%, outperforming seven state-of-the-art (SOTA) models by margins of up to 27.6 percentage points (pp) and delivering a cumulative gain of +10.5 pp over Oriented R-CNN. Full article

Semantic change detection (SCD) effectively captures ground object transition information within change regions, delivering more comprehensive and detailed results than binary change detection (BCD) tasks. The existing multi-task SCD models enable parallel processing of segmentation and BCD of bi-temporal remote sensing images, but they still have shortcomings in feature mining, interaction, and cross-task transfer. To address these limitations, a multi-branch feature interaction network (MBFI-Net) is proposed. MBFI-Net designs parallel encoding branches with attention mechanisms that enhance semantic change perception by jointly modeling global contextual patterns and local details. In addition, MBFI-Net proposes bi-temporal feature interaction (BTFI) and cross-task feature transfer (CTFT) modules to improve feature diversity and representativeness, and combines with prior logical relationship constraints to improve SCD performance. Comparative and ablation studies on the SECOND and Landsat-SCD datasets highlight the superiority and robustness of MBFI-Net, which achieves SeKs of 0.2117 and 0.5543, respectively. Furthermore, MBFI-Net strikes a balance between SCD results and model complexity and has superior detection performance for semantic change categories with a small proportion. Full article

Exogenous fires in underground coal mines are characterized by low illumination, smoke occlusion, heavy dust loading and pseudo fire sources, which jointly degrade image quality and cause missed and false alarms in visual detection. To achieve accurate and real-time early warning under such conditions, this paper proposes a mine exogenous fire detection algorithm based on an improved YOLOv9m, termed PPL-YOLO-F-C. First, a lightweight PP-LCNet backbone is embedded into YOLOv9m to reduce the number of parameters and GFLOPs while maintaining multi-scale feature representation suitable for deployment on resource-constrained edge devices. Second, a Fully Connected Attention (FCAttention) module is introduced to perform fine-grained frequency–channel calibration, enhancing discriminative flame and smoke features and suppressing low-frequency background clutter and non-flame textures. Third, the original upsampling operators in the neck are replaced by the CARAFE content-aware dynamic upsampler to recover blurred flame contours and tenuous smoke edges and to strengthen small-object perception. In addition, an MPDIoU-based bounding-box regression loss is adopted to improve geometric sensitivity and localization accuracy for small fire spots. Experiments on a self-constructed mine fire image dataset comprising 3000 samples show that the proposed PPL-YOLO-F-C model achieves a precision of 97.36%, a recall of 84.91%, mAP@50 of 96.49% and mAP@50:95 of 76.6%, outperforming Faster R-CNN, YOLOv5m, YOLOv7 and YOLOv8m while using fewer parameters and lower computational cost. The results demonstrate that the proposed algorithm provides a robust and efficient solution for real-time exogenous fire detection and edge deployment in complex underground mine environments. Full article

Enhancing low-light images is crucial in computer vision applications. Most existing learning-based models often struggle to balance light enhancement and color correction, while images typically contain different types of information at different levels. Hence, we proposed a multi-scale interactive network with color attention named MSINet to effectively explore these different types of information for lowlight image enhancement (LLIE) tasks. Specifically, the MSINet first employs the CNN-based branch built upon stacked residual channel attention blocks (RCABs) to fully explore the image local features. Meanwhile, the Transformer-based branch constructed by Transformer blocks contains cross-scale attention (CSA) and multi-head self-attention (MHSA) to mine the global features. Notably, the local and global features extracted by each RCAB and Transformer block are interacted with by the fusion module. Additionally, the color correction branch (CCB) based upon self-attention (SA) can learn the color distribution information from the lowlight input for further guaranteeing the color fidelity of the final output. Extensive experiments have demonstrated that our proposed MSINet outperforms state-of-the-art LLIE methods in light enhancement and color correction. Full article

Conveyor belts are among the most critical components of material transport systems across various industrial sectors, including mining, energy, cement production, metallurgy, and logistics. Their reliability directly affects the continuity and operational costs. Traditional methods for assessing belt condition often require downtime, are labor-intensive, and involve a degree of subjectivity. In recent years, there has been a growing interest in non-destructive and remote diagnostic techniques that enable continuous and automated condition monitoring. This paper provides a comprehensive review of current diagnostic solutions, including machine vision systems, infrared thermography, ultrasonic and acoustic techniques, magnetic inspection methods, vibration sensors, and modern approaches based on radar and hyperspectral imaging. Particular attention is paid to the integration of measurement systems with artificial intelligence algorithms for automated damage detection, classification, and failure prediction. The advantages and limitations of each method are discussed, along with the perspectives for future development, such as digital twin concepts and predictive maintenance. The review aims to present recent trends in non-invasive diagnostics of conveyor belts using remote and non-destructive testing techniques, and to identify research directions that can enhance the reliability and efficiency of industrial transport systems. Full article

Underground coal mines are widely distributed across China, and underground mining is highly concealed. The rapid and accurate identification of the spatial distribution of coal mining subsidence over large areas is of significant importance for the reuse of land resources in mining areas and the detection of illegal mining activities. The traditional method of monitoring subsidence basins has limitations in terms of monitoring range and timeliness. The development of synthetic aperture radar (InSAR) technology has provided a valuable tool for monitoring mining subsidence areas. However, this method faces challenges in quickly and effectively monitoring subsidence basins using wide-swath SAR images. With the rapid development of deep learning and computer vision technologies, leveraging advanced deep learning models in combination with InSAR technology has become a crucial research direction to enhance the monitoring efficiency of surface subsidence in mining areas. Therefore, this paper proposes a new method for the rapid identification of mining subsidence basins in mining areas, which integrates Deformable Detection Transformer (Deformable DETR) and InSAR technology. First, the real deformation sample set of the mining area, obtained through interference processing, is combined with simulated deformation samples generated using the dynamic probability integral method, elastic transformation, and various noise synthesis techniques to construct a mixed InSAR sample set. This mixed sample set is then used to train the Deformable DETR model and compared with common deep learning methods. The experimental results show that the monitoring accuracy is significantly improved, with the model achieving a Precision of 0.926, Recall of 0.886, F1-score of 0.905, and mean Intersection over Union (mIoU) of 0.828. The detection model was applied to monitor the dynamically updated mining subsidence in the Huainan mining area from 2023 to 2024, detecting 402 subsidence basins. Further training demonstrates that the model exhibits strong robustness. Therefore, this method reduces the construction cost of the target detection training set and holds significant application potential for monitoring geological disasters in large-scale mining areas. Full article

Underground roadway support is a critical component for ensuring safety in mining operations. In recent years, with the rapid advancement of intelligent technologies, computer vision-based automatic rock bolt detection methods have emerged as a promising alternative to traditional manual inspection. However, the underground mining environment inherently suffers from severely insufficient lighting. Images captured on-site often exhibit problems such as low overall brightness, blurred local details, and severe color distortion. To address the problem, this study proposed a novel low-light image enhancement algorithm, PromptHDR. Based on Transformer architecture, the algorithm effectively suppresses color distortion caused by non-uniform illumination through a Lighting Extraction Module, while simultaneously introducing a Prompt block incorporating a Mamba mechanism to enhance the model’s contextual understanding of the roadway scene and its ability to preserve rock bolt details. Quantitative results demonstrate that the PromptHDR algorithm achieves Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) index scores of 24.19 dB and 0.839, respectively. Furthermore, the enhanced images exhibit more natural visual appearance, adequate brightness recovery, and well-preserved detailed information, establishing a reliable visual foundation for the accurate identification of rock bolts. Full article

Academic dropout is a persistent challenge in higher education, particularly in contexts with socio-economic disparities and diverse learning conditions. Traditional predictive models often fail to capture the complex, non-linear interactions underlying student trajectories due to their reliance on low-dimensional and linear representations. This study introduces a visual representation–based computational approach for a student dropout analysis, applied to a real institutional dataset from Colombia. The methodology transforms structured student records into enriched visual encodings that map variable magnitudes, correlations, and latent relationships into spatial and textural patterns. These image-based representations allow convolutional neural networks (CNNs) to exploit hierarchical feature extraction, uncovering hidden dependencies not accessible through conventional classifiers. Experimental results demonstrate that a Convolutional Neural Network (CNN) trained from scratch outperforms both baseline machine learning models and transfer learning architectures across all evaluation metrics. Beyond predictive accuracy, the approach enhances data expressiveness, interpretability, and generalization, offering a visual-analytical perspective for understanding dropout dynamics. The Colombian case study confirms the feasibility and potential of this strategy in real educational settings, supporting early identification of at-risk students and contributing to the development of robust, explainable models in educational data mining and learning analytics. Full article

Sparse unmixing (SU) has become a research hotspot in hyperspectral image (HSI) analysis in recent years due to its interpretable physical mechanisms and engineering practicality. However, traditional SU methods are confronted with two core bottlenecks: Firstly, the high computational complexity of the abundance matrix inversion severely limits algorithmic efficiency. Secondly, the inherent challenges posed by large-scale highly coherent spectral libraries hinder improvement of unmixing accuracy. To overcome these limitations, this study proposes a novel sub-abundance map regularized sparse unmixing (SARSU) framework based on dynamic abundances subspace awareness. Specifically, first of all, we have developed an intelligent spectral atom selection strategy that employs a designed dynamic activity evaluation mechanism to quantify the participation contribution of spectral library atoms during the unmixing process in real time. This enables adaptive selection of critical subsets to construct active subspace abundance maps, effectively mitigating spectral redundancy interference. Secondly, we innovatively integrated weighted nuclear norm regularization based on sub-abundance maps into the model, deeply mining potential low-rank structures within spatial distribution patterns to significantly enhance the spatial fidelity of unmixing results. Additionally, a multi-directional neighborhood-aware dual total variation (DTV) regularizer was designed, which enforces spatial consistency constraints between adjacent pixels through a four directional (horizontal, vertical, diagonal, and back-diagonal) differential penalty mechanism, ensuring abundance distributions comply with physical diffusion laws of ground objects. Finally, to efficiently solve the proposed objective model, an optimization algorithm based on the Alternating Direction Method of Multipliers (ADMM) was developed. Comparative experiments conducted on two simulated datasets and four real hyperspectral benchmark datasets, alongside comparisons with state-of-the-art methods, validated the efficiency and superiority of the proposed approach. Full article

Aiming at the problems of insufficient environmental perception capability of autonomous mobile robots and low multi-modal data fusion efficiency in the complex underground coal mine environment featuring low illumination, high dust, and dynamic obstacles, a reliable passable region identification method for autonomous mobile robots operating in underground coal mine is proposed in this paper. Through the spatial synchronous installation strategy of dual 4D millimeter-wave radars and dynamic coordinate system registration technology, it increases point cloud density and effectively enhances the spatial characterization of roadway structures and obstacles. Combining the characteristics of infrared thermal imaging and the penetration advantage of millimeter-wave radar, a multi-modal data complementary mechanism based on decision-level fusion is proposed to solve the perceptual blind zones of single sensors in extreme environments. Integrated with lightweight model optimization and system integration technology, an intelligent environmental perception system adaptable to harsh working conditions is constructed. The experiments were carried out in the simulated tunnel. The experiments were carried out in the simulated tunnel. The experimental results indicate that the robot can utilize the data collected by the infrared camera and the radar to identify the specific distance to obstacles, and can smoothly achieve the recognition and marking of passable areas. Full article

Mine rescue robots operate in extremely adverse subterranean environments, where the acquired video data are frequently affected by severe jitter and motion distortion. Such instability leads to the loss of critical visual information, thereby reducing the reliability of rescue decision-making. To address this issue, a dual-channel visual stabilization framework based on the SuperPoint network is proposed, extending the traditional ORB descriptor framework. Here, dual-channel refers to two configurable and mutually exclusive feature extraction paths—an ORB-based path and a SuperPoint-based path—that can be flexibly switched according to scene conditions and computational requirements, rather than operating simultaneously on the same frame. The subsequent stabilization pipeline remains unified and consistent across both modes. The method employs an optimized detector head that integrates deep feature extraction, non-maximum suppression, and boundary filtering to enable precise estimation of inter-frame motion. When combined with smoothing filters, the approach effectively attenuates vibrations induced by irregular terrain and dynamic operational conditions. Experimental evaluations conducted across diverse scenarios demonstrate that the proposed algorithm achieves an average improvement of 27.91% in Peak Signal-to-Noise Ratio (PSNR), a 55.04% reduction in Mean Squared Error (MSE), and more than a twofold increase in the Structural Similarity Index (SSIM) relative to pre-stabilized sequences. Moreover, runtime analysis indicates that the algorithm can operate in near-real-time, supporting its practical deployment on embedded mine rescue robot platforms.These results verify the algorithm’s robustness and applicability in environments requiring high visual stability and image fidelity, providing a reliable foundation for enhanced visual perception and autonomous decision-making in complex disaster scenarios. Full article

The global water sector faces unprecedented challenges from climate change, rapid urbanisation, and ageing infrastructure, necessitating a shift towards proactive, digital strategies. Historically characterised as “data rich but information poor,” the sector struggles with underutilised and siloed operational data. Traditional machine learning (ML) models have provided a foundation for smart water management, and subsequently deep learning (DL) approaches utilising algorithmic breakthroughs and big data have proved to be even more powerful under the right conditions. This paper explores and reviews the transformative potential of Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs), enabling a paradigm shift towards data-centric thinking. GenAI, particularly when augmented with Retrieval-Augmented Generation (RAG) and agentic AI, can create new content, facilitate natural language interaction, synthesise insights from vast unstructured data (of all types including text, images and video) and automate complex, multi-step workflows. Focusing on the critical area of drinking water quality, we demonstrate how these intelligent tools can move beyond reactive systems. A case study is presented which utilises regulatory reports to mine knowledge, providing GenAI-powered chatbots for accessible insights and improved water quality event management. This approach empowers water professionals with dynamic, trustworthy decision support, enhancing the safety and resilience of drinking water supplies by recalling past actions, generating novel insights and simulating response scenarios. Full article