MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining area
Ground fissures caused by high-intensity underground coal mining activities will damage the ecological environment and endanger mine safety. The complicated surface conditions in coal mining areas make fissure detection labor-intensive work, and manual fissure extraction is inefficient, restricting...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-11-01
|
| Series: | International Journal of Applied Earth Observations and Geoinformation |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1569843222002278 |
| _version_ | 1811243578055196672 |
|---|---|
| author | Xiao Jiang Shanjun Mao Mei Li Hui Liu Haoyuan Zhang Shuwei Fang Mingze Yuan Chi Zhang |
| author_facet | Xiao Jiang Shanjun Mao Mei Li Hui Liu Haoyuan Zhang Shuwei Fang Mingze Yuan Chi Zhang |
| author_sort | Xiao Jiang |
| collection | DOAJ |
| description | Ground fissures caused by high-intensity underground coal mining activities will damage the ecological environment and endanger mine safety. The complicated surface conditions in coal mining areas make fissure detection labor-intensive work, and manual fissure extraction is inefficient, restricting the surface monitoring work. In this study, an efficient Deep Learning (DL) network named MFPA-Net (Multi-scale Feature Pyramids and Attention Network) is proposed to extract ground fissures in Unmanned Aerial Vehicle (UAV) images from the coal mining area automatically and accurately. In MFPA-Net, the Dilated Residual Networks (DRN) are used to extract diverse context information, the Dual Attention Mechanism (DAM) is introduced to integrate the dependence of pixels' spatial location and feature channels to generate high-level features, the Atrous Spatial Pyramid Pooling (ASPP) is utilized for mining multi-scale context information from high-level features, and the Multi-scale Feature Pyramid Network (MFPN) is designed to combine high-level and low-level features. Moreover, the Focal Tversky Loss Function is adopted to handle the unbalanced samples. To promote DL technologies application in the fissure monitoring of mining areas, the GFCMA (Ground Fissures of the Coal Mining Area) dataset is constructed. Experiments on GFCMA show that MFPA-Net can achieve high Precision (69.4%), Recall (70.7%), F1-Score (70.0%), and Mean Intersection over Union (MIoU) (75.1%) simultaneously, which significantly outperform traditional image processing methods and recently DL networks. Experiments on public pavement datasets Crack500 and DeepCrack prove MFPA-Net's high reliability and widespread applicability. The performance of the trained MFPA-Net on real large-scale scenarios demonstrates its practical value, strong robustness, and high efficiency. This study provides a solution for rapid monitoring of ground fissures under complicated surface conditions, which can serve the safe production and ecological restoration high-efficiently in mining areas. |
| first_indexed | 2024-04-12T14:09:47Z |
| format | Article |
| id | doaj.art-7869650c8c5540ddbb33e9f7e4779782 |
| institution | Directory Open Access Journal |
| issn | 1569-8432 |
| language | English |
| last_indexed | 2024-04-12T14:09:47Z |
| publishDate | 2022-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Applied Earth Observations and Geoinformation |
| spelling | doaj.art-7869650c8c5540ddbb33e9f7e47797822022-12-22T03:29:55ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322022-11-01114103039MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining areaXiao Jiang0Shanjun Mao1Mei Li2Hui Liu3Haoyuan Zhang4Shuwei Fang5Mingze Yuan6Chi Zhang7School of Earth and Space Sciences, Peking University, Beijing 100871, ChinaSchool of Earth and Space Sciences, Peking University, Beijing 100871, China; Corresponding author.School of Earth and Space Sciences, Peking University, Beijing 100871, ChinaSchool of Remote Sensing & Geomatics Engineering, Nanjing University of Information Science & Technology, Nanjing 210044, ChinaSchool of Earth and Space Sciences, Peking University, Beijing 100871, ChinaSchool of Earth and Space Sciences, Peking University, Beijing 100871, ChinaSchool of Earth and Space Sciences, Peking University, Beijing 100871, ChinaNorthern Shaanxi Mining Zhangjiamao Co. Ltd., Shenmu 719300, ChinaGround fissures caused by high-intensity underground coal mining activities will damage the ecological environment and endanger mine safety. The complicated surface conditions in coal mining areas make fissure detection labor-intensive work, and manual fissure extraction is inefficient, restricting the surface monitoring work. In this study, an efficient Deep Learning (DL) network named MFPA-Net (Multi-scale Feature Pyramids and Attention Network) is proposed to extract ground fissures in Unmanned Aerial Vehicle (UAV) images from the coal mining area automatically and accurately. In MFPA-Net, the Dilated Residual Networks (DRN) are used to extract diverse context information, the Dual Attention Mechanism (DAM) is introduced to integrate the dependence of pixels' spatial location and feature channels to generate high-level features, the Atrous Spatial Pyramid Pooling (ASPP) is utilized for mining multi-scale context information from high-level features, and the Multi-scale Feature Pyramid Network (MFPN) is designed to combine high-level and low-level features. Moreover, the Focal Tversky Loss Function is adopted to handle the unbalanced samples. To promote DL technologies application in the fissure monitoring of mining areas, the GFCMA (Ground Fissures of the Coal Mining Area) dataset is constructed. Experiments on GFCMA show that MFPA-Net can achieve high Precision (69.4%), Recall (70.7%), F1-Score (70.0%), and Mean Intersection over Union (MIoU) (75.1%) simultaneously, which significantly outperform traditional image processing methods and recently DL networks. Experiments on public pavement datasets Crack500 and DeepCrack prove MFPA-Net's high reliability and widespread applicability. The performance of the trained MFPA-Net on real large-scale scenarios demonstrates its practical value, strong robustness, and high efficiency. This study provides a solution for rapid monitoring of ground fissures under complicated surface conditions, which can serve the safe production and ecological restoration high-efficiently in mining areas.http://www.sciencedirect.com/science/article/pii/S1569843222002278Ground fissures extractionDeep learningSemantic segmentationUAV image processingCoal mining areas |
| spellingShingle | Xiao Jiang Shanjun Mao Mei Li Hui Liu Haoyuan Zhang Shuwei Fang Mingze Yuan Chi Zhang MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining area International Journal of Applied Earth Observations and Geoinformation Ground fissures extraction Deep learning Semantic segmentation UAV image processing Coal mining areas |
| title | MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining area |
| title_full | MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining area |
| title_fullStr | MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining area |
| title_full_unstemmed | MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining area |
| title_short | MFPA-Net: An efficient deep learning network for automatic ground fissures extraction in UAV images of the coal mining area |
| title_sort | mfpa net an efficient deep learning network for automatic ground fissures extraction in uav images of the coal mining area |
| topic | Ground fissures extraction Deep learning Semantic segmentation UAV image processing Coal mining areas |
| url | http://www.sciencedirect.com/science/article/pii/S1569843222002278 |
| work_keys_str_mv | AT xiaojiang mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea AT shanjunmao mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea AT meili mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea AT huiliu mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea AT haoyuanzhang mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea AT shuweifang mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea AT mingzeyuan mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea AT chizhang mfpanetanefficientdeeplearningnetworkforautomaticgroundfissuresextractioninuavimagesofthecoalminingarea |