High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny Algorithm
Realizing high-precision real-time underwater detection has been a pressing issue for intelligent underwater robots in recent years. Poor quality of underwater datasets leads to low accuracy of detection models. To handle this problem, an improved YOLOV4-Tiny algorithm is proposed. The CSPrestblock_...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Journal of Marine Science and Engineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-1312/10/12/1821 |
| _version_ | 1827638229664268288 |
|---|---|
| author | Kun Yu Yufeng Cheng Zhuangtao Tian Kaihua Zhang |
| author_facet | Kun Yu Yufeng Cheng Zhuangtao Tian Kaihua Zhang |
| author_sort | Kun Yu |
| collection | DOAJ |
| description | Realizing high-precision real-time underwater detection has been a pressing issue for intelligent underwater robots in recent years. Poor quality of underwater datasets leads to low accuracy of detection models. To handle this problem, an improved YOLOV4-Tiny algorithm is proposed. The CSPrestblock_body in YOLOV4-Tiny is replaced with Ghostblock_body, which is stacked by Ghost modules in the CSPDarknet53-Tiny backbone network to reduce the computation complexity. The convolutional block attention module (CBAM) is integrated to the algorithm in order to find the attention region in scenarios with dense objects. Then, underwater data is effectively improved by combining the Instance-Balanced Augmentation, underwater image restoration, and Mosaic algorithm. Finally, experiments demonstrate that the YOLOV4-Tinier has a mean Average Precision (mAP) of 80.77% on the improved underwater dataset and a detection speed of 86.96 fps. Additionally, compared to the baseline model YOLOV4-Tiny, YOLOV4-Tinier reduces about model size by about 29%, which is encouraging and competitive. |
| first_indexed | 2024-03-09T16:14:37Z |
| format | Article |
| id | doaj.art-249de2452a0c4d838ef56add1e018c20 |
| institution | Directory Open Access Journal |
| issn | 2077-1312 |
| language | English |
| last_indexed | 2024-03-09T16:14:37Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Marine Science and Engineering |
| spelling | doaj.art-249de2452a0c4d838ef56add1e018c202023-11-24T15:54:53ZengMDPI AGJournal of Marine Science and Engineering2077-13122022-11-011012182110.3390/jmse10121821High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny AlgorithmKun Yu0Yufeng Cheng1Zhuangtao Tian2Kaihua Zhang3Henan Key Laboratory of Infrared Materials & Spectrum Measures and Applications, School of Physics, Henan Normal University, Xinxiang 453007, ChinaHenan Key Laboratory of Infrared Materials & Spectrum Measures and Applications, School of Physics, Henan Normal University, Xinxiang 453007, ChinaHenan Key Laboratory of Infrared Materials & Spectrum Measures and Applications, School of Physics, Henan Normal University, Xinxiang 453007, ChinaHenan Key Laboratory of Infrared Materials & Spectrum Measures and Applications, School of Physics, Henan Normal University, Xinxiang 453007, ChinaRealizing high-precision real-time underwater detection has been a pressing issue for intelligent underwater robots in recent years. Poor quality of underwater datasets leads to low accuracy of detection models. To handle this problem, an improved YOLOV4-Tiny algorithm is proposed. The CSPrestblock_body in YOLOV4-Tiny is replaced with Ghostblock_body, which is stacked by Ghost modules in the CSPDarknet53-Tiny backbone network to reduce the computation complexity. The convolutional block attention module (CBAM) is integrated to the algorithm in order to find the attention region in scenarios with dense objects. Then, underwater data is effectively improved by combining the Instance-Balanced Augmentation, underwater image restoration, and Mosaic algorithm. Finally, experiments demonstrate that the YOLOV4-Tinier has a mean Average Precision (mAP) of 80.77% on the improved underwater dataset and a detection speed of 86.96 fps. Additionally, compared to the baseline model YOLOV4-Tiny, YOLOV4-Tinier reduces about model size by about 29%, which is encouraging and competitive.https://www.mdpi.com/2077-1312/10/12/1821underwater biological detectionYOLOV4-Tinydata augmentationimage restoration |
| spellingShingle | Kun Yu Yufeng Cheng Zhuangtao Tian Kaihua Zhang High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny Algorithm Journal of Marine Science and Engineering underwater biological detection YOLOV4-Tiny data augmentation image restoration |
| title | High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny Algorithm |
| title_full | High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny Algorithm |
| title_fullStr | High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny Algorithm |
| title_full_unstemmed | High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny Algorithm |
| title_short | High Speed and Precision Underwater Biological Detection Based on the Improved YOLOV4-Tiny Algorithm |
| title_sort | high speed and precision underwater biological detection based on the improved yolov4 tiny algorithm |
| topic | underwater biological detection YOLOV4-Tiny data augmentation image restoration |
| url | https://www.mdpi.com/2077-1312/10/12/1821 |
| work_keys_str_mv | AT kunyu highspeedandprecisionunderwaterbiologicaldetectionbasedontheimprovedyolov4tinyalgorithm AT yufengcheng highspeedandprecisionunderwaterbiologicaldetectionbasedontheimprovedyolov4tinyalgorithm AT zhuangtaotian highspeedandprecisionunderwaterbiologicaldetectionbasedontheimprovedyolov4tinyalgorithm AT kaihuazhang highspeedandprecisionunderwaterbiologicaldetectionbasedontheimprovedyolov4tinyalgorithm |