An underground coal mine multi-target detection algorithm
Currently, underground coal mine target detection algorithms based on deep learning show poor performance in detecting complex small targets under conditions of uneven light intensity distribution, complex target environments, and imbalanced multi-class target scale distribution, often resulting in...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Industry and Mine Automation
2024-12-01
|
| Series: | Gong-kuang zidonghua |
| Subjects: | |
| Online Access: | http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.2024090035 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832591077132992512 |
|---|---|
| author | FAN Shoujun CHEN Xilin WEI Liangyue WANG Qingyu ZHANG Shiyuan DONG Fei LEI Shaohua |
| author_facet | FAN Shoujun CHEN Xilin WEI Liangyue WANG Qingyu ZHANG Shiyuan DONG Fei LEI Shaohua |
| author_sort | FAN Shoujun |
| collection | DOAJ |
| description | Currently, underground coal mine target detection algorithms based on deep learning show poor performance in detecting complex small targets under conditions of uneven light intensity distribution, complex target environments, and imbalanced multi-class target scale distribution, often resulting in missed detection and false detection. To address these issues, based on the single-stage target detection algorithm YOLOv8n, this study proposed an underground coal mine multi-target detection algorithm based on feature extraction by dynamic snake convolution (FEDSC)-feature fusion by bi-directional feature pyramid network and semantic and detail fusion (FFBD). FEDSC replaced the backbone network of YOLOv8n to expand the receptive field, while FFBD acted as the neck network to reduce target false detection and missed detection. Additionally, a decoupling detection head of SIoU was used as the detection layer to improve the model's adaptability to small targets and the convergence speed. The results showed that: ① The mAP@0.5 of the FEDSC-FFBD algorithm was 97.00%, the number of model parameters was 4.22×106, and the number of floating point operations per second was 21.7×109. ② The mAP@0.5 of the FEDSC-FFBD alorithm was 3.40% higher than the YOLOv8n algorithm, and the recognition accuracy of the helmet small target was 90.90%, 11% higher than the YOLOv8n algorithm. ③ Compared with other YOLO series algorithms, the FEDSC-FFBD algorithm achieved the highest mAP@0.5, which was 3.60%, 1%, 10.50%, and 6.40% higher than YOLOv5s, YOLOv9c, YOLOv10n, and YOLOv11n algorithms, respectively. ④ The FEDSC-FFBD algorithm improved the detection accuracy of multi-class targets and reduced missed detection and false detection of small targets under conditions of uneven light intensity distribution, complex target environments, and imbalanced target scale distribution in underground coal mine. The underground coal mine multi-target detection algorithm based on FEDSC-FFBD overcame the challenge of small-scale target detection caused by uneven light intensity distribution without relying on image quality enhancement algorithms. |
| format | Article |
| id | doaj-art-0cabc0c8f8124945858983e94edb23b9 |
| institution | Kabale University |
| issn | 1671-251X |
| language | zho |
| publishDate | 2024-12-01 |
| publisher | Editorial Department of Industry and Mine Automation |
| record_format | Article |
| series | Gong-kuang zidonghua |
| spelling | doaj-art-0cabc0c8f8124945858983e94edb23b92025-01-23T02:17:44ZzhoEditorial Department of Industry and Mine AutomationGong-kuang zidonghua1671-251X2024-12-01501217318210.13272/j.issn.1671-251x.2024090035An underground coal mine multi-target detection algorithmFAN Shoujun0CHEN Xilin1WEI Liangyue2WANG Qingyu3ZHANG Shiyuan4DONG FeiLEI Shaohua5Yankuang Energy Group Company Limited, Jining 273500, ChinaSchool of Information and Control Engineering, China University of Mining and Technology, Xuzhou 221116, ChinaYankuang Energy Group Company Limited, Jining 273500, ChinaYankuang Energy Group Company Limited, Jining 273500, ChinaSchool of Information and Control Engineering, China University of Mining and Technology, Xuzhou 221116, ChinaXuzhou High-tech Zone Safety Emergency Equipment Industry Technology Research Institute, Xuzhou 221008, ChinaCurrently, underground coal mine target detection algorithms based on deep learning show poor performance in detecting complex small targets under conditions of uneven light intensity distribution, complex target environments, and imbalanced multi-class target scale distribution, often resulting in missed detection and false detection. To address these issues, based on the single-stage target detection algorithm YOLOv8n, this study proposed an underground coal mine multi-target detection algorithm based on feature extraction by dynamic snake convolution (FEDSC)-feature fusion by bi-directional feature pyramid network and semantic and detail fusion (FFBD). FEDSC replaced the backbone network of YOLOv8n to expand the receptive field, while FFBD acted as the neck network to reduce target false detection and missed detection. Additionally, a decoupling detection head of SIoU was used as the detection layer to improve the model's adaptability to small targets and the convergence speed. The results showed that: ① The mAP@0.5 of the FEDSC-FFBD algorithm was 97.00%, the number of model parameters was 4.22×106, and the number of floating point operations per second was 21.7×109. ② The mAP@0.5 of the FEDSC-FFBD alorithm was 3.40% higher than the YOLOv8n algorithm, and the recognition accuracy of the helmet small target was 90.90%, 11% higher than the YOLOv8n algorithm. ③ Compared with other YOLO series algorithms, the FEDSC-FFBD algorithm achieved the highest mAP@0.5, which was 3.60%, 1%, 10.50%, and 6.40% higher than YOLOv5s, YOLOv9c, YOLOv10n, and YOLOv11n algorithms, respectively. ④ The FEDSC-FFBD algorithm improved the detection accuracy of multi-class targets and reduced missed detection and false detection of small targets under conditions of uneven light intensity distribution, complex target environments, and imbalanced target scale distribution in underground coal mine. The underground coal mine multi-target detection algorithm based on FEDSC-FFBD overcame the challenge of small-scale target detection caused by uneven light intensity distribution without relying on image quality enhancement algorithms.http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.2024090035multi-target detection in underground coal mineyolov8ndynamic serpentine convolutionca attention mechanismfeature extractionfeature fusion |
| spellingShingle | FAN Shoujun CHEN Xilin WEI Liangyue WANG Qingyu ZHANG Shiyuan DONG Fei LEI Shaohua An underground coal mine multi-target detection algorithm Gong-kuang zidonghua multi-target detection in underground coal mine yolov8n dynamic serpentine convolution ca attention mechanism feature extraction feature fusion |
| title | An underground coal mine multi-target detection algorithm |
| title_full | An underground coal mine multi-target detection algorithm |
| title_fullStr | An underground coal mine multi-target detection algorithm |
| title_full_unstemmed | An underground coal mine multi-target detection algorithm |
| title_short | An underground coal mine multi-target detection algorithm |
| title_sort | underground coal mine multi target detection algorithm |
| topic | multi-target detection in underground coal mine yolov8n dynamic serpentine convolution ca attention mechanism feature extraction feature fusion |
| url | http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.2024090035 |
| work_keys_str_mv | AT fanshoujun anundergroundcoalminemultitargetdetectionalgorithm AT chenxilin anundergroundcoalminemultitargetdetectionalgorithm AT weiliangyue anundergroundcoalminemultitargetdetectionalgorithm AT wangqingyu anundergroundcoalminemultitargetdetectionalgorithm AT zhangshiyuan anundergroundcoalminemultitargetdetectionalgorithm AT dongfei anundergroundcoalminemultitargetdetectionalgorithm AT leishaohua anundergroundcoalminemultitargetdetectionalgorithm AT fanshoujun undergroundcoalminemultitargetdetectionalgorithm AT chenxilin undergroundcoalminemultitargetdetectionalgorithm AT weiliangyue undergroundcoalminemultitargetdetectionalgorithm AT wangqingyu undergroundcoalminemultitargetdetectionalgorithm AT zhangshiyuan undergroundcoalminemultitargetdetectionalgorithm AT dongfei undergroundcoalminemultitargetdetectionalgorithm AT leishaohua undergroundcoalminemultitargetdetectionalgorithm |