Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer Features
Fish object detection has attracted significant attention because of the considerable role that fish play in human society and ecosystems and the necessity to gather more comprehensive fish data through underwater videos or images. However, fish detection has always faced difficulties with the occlu...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/23/12645 |
| _version_ | 1827592489391882240 |
|---|---|
| author | Enze Li Qibiao Wang Jinzhao Zhang Weihan Zhang Hanlin Mo Yadong Wu |
| author_facet | Enze Li Qibiao Wang Jinzhao Zhang Weihan Zhang Hanlin Mo Yadong Wu |
| author_sort | Enze Li |
| collection | DOAJ |
| description | Fish object detection has attracted significant attention because of the considerable role that fish play in human society and ecosystems and the necessity to gather more comprehensive fish data through underwater videos or images. However, fish detection has always faced difficulties with the occlusion problem because of dense populations and underwater plants that obscure them, and no perfect solution has been found until now. To address the occlusion issue in fish detection, the following effort was made: creating a dataset of occluded fishes, integrating the innovative modules in Real-time Detection Transformer (RT-DETR) into You Only Look Once v8 (YOLOv8), and applying repulsion loss. The results show that in the occlusion dataset, the mAP of the original YOLOv8 is 0.912, while the mAP of our modified YOLOv8 is 0.971. In addition, our modified YOLOv8 also has better performance than the original YOLOv8 in terms of loss curves, F1–Confidence curves, P–R curves, the mAP curve and the actual detection effects. All these indicate that our modified YOLOv8 is suitable for fish detection in occlusion scenes. |
| first_indexed | 2024-03-09T01:56:14Z |
| format | Article |
| id | doaj.art-e6c1a34dc9094c1a850a644f98f5ee47 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T01:56:14Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-e6c1a34dc9094c1a850a644f98f5ee472023-12-08T15:11:15ZengMDPI AGApplied Sciences2076-34172023-11-0113231264510.3390/app132312645Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer FeaturesEnze Li0Qibiao Wang1Jinzhao Zhang2Weihan Zhang3Hanlin Mo4Yadong Wu5School of Computer Science and Engineering, Sichuan University of Science and Engineering, Zigong 643000, ChinaSchool of Computer Science and Engineering, Sichuan University of Science and Engineering, Zigong 643000, ChinaThird Institute of Oceanography, Ministry of Natural Resources, Xiamen 361000, ChinaSchool of Computer Science and Engineering, Sichuan University of Science and Engineering, Zigong 643000, ChinaSchool of Computer Science and Engineering, Sichuan University of Science and Engineering, Zigong 643000, ChinaSchool of Physics and Electronic Engineering, Sichuan University of Science and Engineering, Zigong 643000, ChinaFish object detection has attracted significant attention because of the considerable role that fish play in human society and ecosystems and the necessity to gather more comprehensive fish data through underwater videos or images. However, fish detection has always faced difficulties with the occlusion problem because of dense populations and underwater plants that obscure them, and no perfect solution has been found until now. To address the occlusion issue in fish detection, the following effort was made: creating a dataset of occluded fishes, integrating the innovative modules in Real-time Detection Transformer (RT-DETR) into You Only Look Once v8 (YOLOv8), and applying repulsion loss. The results show that in the occlusion dataset, the mAP of the original YOLOv8 is 0.912, while the mAP of our modified YOLOv8 is 0.971. In addition, our modified YOLOv8 also has better performance than the original YOLOv8 in terms of loss curves, F1–Confidence curves, P–R curves, the mAP curve and the actual detection effects. All these indicate that our modified YOLOv8 is suitable for fish detection in occlusion scenes.https://www.mdpi.com/2076-3417/13/23/12645fish detectionocclusionobject detectionYOLOv8RT-DETR |
| spellingShingle | Enze Li Qibiao Wang Jinzhao Zhang Weihan Zhang Hanlin Mo Yadong Wu Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer Features Applied Sciences fish detection occlusion object detection YOLOv8 RT-DETR |
| title | Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer Features |
| title_full | Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer Features |
| title_fullStr | Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer Features |
| title_full_unstemmed | Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer Features |
| title_short | Fish Detection under Occlusion Using Modified You Only Look Once v8 Integrating Real-Time Detection Transformer Features |
| title_sort | fish detection under occlusion using modified you only look once v8 integrating real time detection transformer features |
| topic | fish detection occlusion object detection YOLOv8 RT-DETR |
| url | https://www.mdpi.com/2076-3417/13/23/12645 |
| work_keys_str_mv | AT enzeli fishdetectionunderocclusionusingmodifiedyouonlylookoncev8integratingrealtimedetectiontransformerfeatures AT qibiaowang fishdetectionunderocclusionusingmodifiedyouonlylookoncev8integratingrealtimedetectiontransformerfeatures AT jinzhaozhang fishdetectionunderocclusionusingmodifiedyouonlylookoncev8integratingrealtimedetectiontransformerfeatures AT weihanzhang fishdetectionunderocclusionusingmodifiedyouonlylookoncev8integratingrealtimedetectiontransformerfeatures AT hanlinmo fishdetectionunderocclusionusingmodifiedyouonlylookoncev8integratingrealtimedetectiontransformerfeatures AT yadongwu fishdetectionunderocclusionusingmodifiedyouonlylookoncev8integratingrealtimedetectiontransformerfeatures |