International Journal of New Developments in Engineering and Society, 2025, 9(2); doi: 10.25236/IJNDES.2025.090213.
Yuan Liu1, Yang Xu1, Xingkun Li2, Guodong Chen1
1School of Naval Architecture and Port Engineering, Shandong Jiao Tong University, Weihai, Shandong Province, 264209, China
2School of Physical Sciences, Qingdao University, Qingdao, Shandong Province, 266071, China
Underwater environments present unique and challenging conditions for target detection. Factors such as light attenuation, suspended particles, and water turbulence contribute to significant color distortion, low contrast, complex target morphology, and dense background clutter. To address these issues, this paper proposes UWYOLO, a specialized model based on the YOLOv8 framework that incorporates the Large Separable Kernel Attention (LSKA) module. Designed to enhance detection performance in such difficult conditions, UWYOLO improves the extraction of shape features and increases robustness, particularly for detecting small targets in cluttered backgrounds. By emphasizing shape information over texture, the model also demonstrates improved applicability in practical underwater tasks such as biodiversity monitoring and environmental surveys. Experimental results show that UWYOLO achieves a mean average precision (mAP) of 84.9% on a specialized underwater dataset, exceeding the baseline YOLOv8 by 1.0%, confirming its advanced accuracy in underwater target detection.
Underwater object detection; Large separable kernel attention; Deep learning; Shape feature extraction; UWYOLO
Yuan Liu, Yang Xu, Xingkun Li, Guodong Chen. Research on Underwater Object Detection Method Based on Deep Learning. International Journal of New Developments in Engineering and Society (2025), Vol. 9, Issue 2: 78-83. https://doi.org/10.25236/IJNDES.2025.090213.
[1] Jia J, Fu M, Liu X, et al. Underwater object detection based on improved efficientdet[J]. Remote Sensing, 2022, 14(18): 4487.
[2] Jian M, Liu X, Luo H, et al. Underwater image processing and analysis: A review[J]. Signal Processing: Image Communication, 2021, 91: 116088.
[3] Li C, Guo J. Underwater image enhancement by dehazing and color correction[J]. Journal of Electronic Imaging, 2015, 24(3): 033023.
[4] Ancuti C O, Ancuti C, De Vleeschouwer C, et al. Color balance and fusion for underwater image enhancement[J]. IEEE Transactions on image processing, 2017, 27(1): 379-393.
[5] Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox detector[C]//European conference on computer vision. Cham: Springer International Publishing, 2016: 21-37.
[6] Bochkovskiy A, Wang C Y, Liao H Y M. Yolov4: Optimal speed and accuracy of object detection[J]. arxiv preprint arxiv:2004.10934, 2020.
[7] Redmon J, Farhadi A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 7263-7271.
[8] Redmon J, Farhadi A. Yolov3: An incremental improvement[J]. arxiv preprint arxiv:1804.02767, 2018.
[9] Wang C, He W, Nie Y, et al. Gold-YOLO: Efficient object detector via gather-and-distribute mechanism[J]. Advances in Neural Information Processing Systems, 2023, 36: 51094-51112.
[10] Girshick R. Fast r-cnn[C]//Proceedings of the IEEE international conference on computer vision. 2015: 1440-1448.
[11] Ren S, He K, Girshick R, et al. Faster r-cnn: Towards real-time object detection with region proposal networks[J]. Advances in neural information processing systems, 2015, 28.
[12] Liu K, Peng L, Tang S. Underwater object detection using TC-YOLO with attention mechanisms[J]. Sensors, 2023, 23(5): 2567.
[13] Luo S, Dong C, Dong G, et al. YOLO-DAFS: A Composite-Enhanced Underwater Object Detection Algorithm[J]. Journal of Marine Science and Engineering, 2025, 13(5): 947.
[14] Hu K, Lu F, Lu M, et al. A marine object detection algorithm based on SSD and feature enhancement[J]. Complexity, 2020, 2020(1): 5476142.
[15] Hu J H. Object Detection Model for Marine Organisms Based on Faster R-CNN[J]. Advances in Engineering Technology Research, 2024, 9(1): 567-567.
[16] Liu J, Liu S, Xu S, et al. Two-stage underwater object detection network using swin transformer[J]. IEEE Access, 2022, 10: 117235-117247.
[17] Lau K W, Po L M, Rehman Y A U. Large separable kernel attention: Rethinking the large kernel attention design in cnn[J]. Expert Systems with Applications, 2024, 236: 121352.