Research Review of Underwater Target Detection Technology based on Path Aggregation Network

Peihan Yang; Haifeng Yu; Zonglin Li; Liangjing Fan; Wenzewei Liu

doi:10.6919/ICJE.202505_11(5).0013

Authors

Peihan Yang
Haifeng Yu
Zonglin Li
Liangjing Fan
Wenzewei Liu

DOI:

https://doi.org/10.6919/ICJE.202505_11(5).0013

Keywords:

Underwater Target Detection; Path Aggregation Network; YOLO; Attention Mechanism.

Abstract

Underwater target detection technology is of great significance in marine research and applications, but its performance is severely constrained by the complexity of the underwater environment, such as light attenuation, scattering and degradation of image quality due to turbid water. Traditional methods are limited in their effectiveness due to the difficulty in dealing with multi-scale targets and complex backgrounds. In recent years, deep learning-based path aggregation network (PANet), as an efficient feature fusion structure, is able to effectively fuse multi-scale features and enhance the target detection performance through top-down and bottom-up path design. This paper systematically reviews the technical principles of PANet and its current research status in underwater target detection, then focuses on the analysis of the improved PANet model based on YOLO series, and discusses the challenges faced by this technology in the directions of model lightweighting, data diversity enhancement and multimodal fusion. Future research needs to further optimize the computational efficiency, expand the dataset size and explore the cross-modal data fusion, in order to promote the practical application of underwater target detection technology in complex ocean scenarios.

Downloads

Download data is not yet available.

References

[1] Lu Jiahao. Research on super-resolution algorithm of underwater image based on improved channel attention residual network [D]. Guilin University of Electronic Science and Technology,2023.

[2] Qian Junbing, Jiang Liangwei, Quan Haiyan, et al. An underwater image enhancement method combining multi-channel equalization and multi-scale fusion [J]. Optoelectronics. Laser, 2022, 33(4):10.

[3] Dhillon A, Verma G K. Convolutional neural network: a review of models, methodologies and applications to object detection[J]. Progress in Artificial Intelligence, 2020, 9(2): 85-112.

[4] Liu S, Qi L, Qin H, et al. Path aggregation network for instance segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 8759-8768.

[5] Lin T Y, Dollar P, Girshick R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 2117-2125.

[6] Chen Mengxing, Chen Wei, Chen Guojun, et al. Lightweight Algorithm of Underwater human sonar image real-time target Detection based on YOLOv8 [J]. Light Industry Science and Technology, 2024, 40 (06): 111-114.

[7] Liu T, Pang B, Ai S, et al. Study on visual detection algorithm of sea surface targets based on improved YOLOv3[J]. Sensors, 2020, 20(24): 7263.

[8] Liu J, Liu S, Xu S, et al. Two-stage underwater object detection network using swin transformer[J]. IEEE Access, 2022, 10: 117235-117247.

[9] Zhang Y, Liu T, Yu X. Contextual and lightweight network for underwater object detection with self-attention mechanism[C]//2023 IEEE International Conference on Mechatronics and Automation (ICMA). IEEE, 2023: 1644-1649.

[10] Zhang M, Xu S, Song W, et al. Lightweight underwater object detection based on yolo v4 and multi-scale attentional feature fusion[J]. Remote Sensing, 2021, 13(22): 4706.

[11] Yu H, Li X, Feng Y, et al. Multiple attentional path aggregation network for marine object detection[J]. Applied intelligence, 2023, 53(2): 2434-2451.

[12] Zou Ziyin, Gai Shaoyan, Da Feipeng, et al. Acta Optica Sinica, 2021, 41(15): 1515001.] Acta Optica Sinica, 2013, 41(15): 1515001.

[13] Chi Y, Li J, Fan H. Pyramid-attention based multi-scale feature fusion network for multispectral pan-sharpening[J]. Applied Intelligence, 2022, 52(5): 5353-5365.

[14] Li Z, Xie Z, Duan P, et al. Dual spatial attention network for underwater object detection with sonar imagery[J]. IEEE Sensors Journal, 2024, 24(5): 6998-7008.

[15] He D, Li Y, Chen L, et al. A two-stage image enhancement and dynamic feature aggregation framework for gastroscopy image segmentation[J]. Neurocomputing, 2024, 601: 128228.

[16] Dang L, Huangfu P, Hou Y, et al. A path aggregation network based on residual feature enhancement for object detection in remote sensing imagery[J]. Remote Sensing Letters, 2023, 14(6): 598-608.

[17] Fang S, Zhang B, Hu J. Improved mask R-CNN multi-target detection and segmentation for autonomous driving in complex scenes[J]. Sensors, , 2023, 23(8): 3853.

[18] Fu C, Liu R, Fan X, et al. Rethinking general underwater object detection: Datasets, challenges, and solutions[J]. Neurocomputing, 2023, 517: 243-256.

[19] Liu C, Li H, Wang S, et al. A dataset and benchmark of underwater object detection for robot picking[C]//2021 IEEE international conference on multimedia & expo workshops (ICMEW). IEEE, 2021: 1-6.

[20] Cai S, Li G, Shan Y. Underwater object detection using collaborative weakly supervision[J]. Computers and Electrical Engineering, 2022, 102: 108159.

[21] Fish4Knowledge: collecting and analyzing massive coral reef fish video data[M]. Berlin/Heidelberg, Germany: Springer, 2016.

[22] Hong J, Fulton M, Sattar J. Trashcan: A semantically-segmented dataset towards visual detection of marine debris[J]. arXiv preprint arXiv:2007.08097, 2020.

[23] Islam M J, Edge C, Xiao Y, et al. Semantic segmentation of underwater imagery: Dataset and benchmark[C]//2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2020: 1769-1776.

[24] Lu H, Uemura T, Wang D, et al. Deep-sea organisms tracking using dehazing and deep learning[J]. Mobile Networks and Applications, 2020, 25: 1008-1015.

[25] Jiang P, Ergu D, Liu F, et al. A Review of Yolo algorithm developments[J]. Procedia computer science, 2022, 199: 1066-1073.

[26] Hu X, Liu Y, Zhao Z, et al. Real-time detection of uneaten feed pellets in underwater images for aquaculture using an improved YOLO-V4 network[J]. Computers and electronics in agriculture, 2021, 185: 106135.

[27] Lei F, Tang F, Li S. Underwater target detection algorithm based on improved YOLOv5[J]. Journal of Marine Science and Engineering, 2022, 10(3): 310.

[28] Zhang J, Yongpan W, Xianchong X, et al. YoloXT: A object detection algorithm for marine benthos[J]. Ecological Informatics, 2022, 72: 101923.

[29] Chen L, Yin T, Zhou S, et al. Underwater target detection algorithm based on feature fusion enhancement[J]. Electronics, 2023, 12(13): 2756.

[30] Zhao Y, Sun F, Wu X. FEB-YOLOv8: A multi-scale lightweight detection model for underwater object detection[J]. PLoS One, 2024, 19(9): e0311173.

[31] Tian Y, Liu Y, Lin B, et al. Research on marine flexible biological target detection based on improved YOLOv8 algorithm[J]. PeerJ Computer Science, 2024, 10: e2271.

[32] Meng Xiao-Xuan. Research on object detection Algorithm based on deep Active Weakly Supervised Learning [J]. Information and Computer (Theoretical Edition), 2021, 33 (02): 61-64.]

[33] Zech J R, Badgeley M A, Liu M, et al. Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: a cross-sectional study[J]. PLoS medicine, 2018, 15(11): e1002683.

[34] Zhang C, Bengio S, Hardt M, et al. Understanding deep learning (still) requires rethinking generalization[J]. Communications of the ACM, 2021, 64(3): 107-115.

[35] Li Ming-gui, Zhou Huan-yin, Gong Li-wen. Application of ROV underwater obstacle detection and obstacle avoidance technology [J]. Computer Engineering and Applications, 2024, 60 (17): 34-47.

[36] Chen H, Wang Z, Qin H, et al. UAMFDet: Acoustic‐Optical Fusion for Underwater Multi‐Modal Object Detection[J]. Journal of Field Robotics, 2024.

[37] Lahat D, Adali T, Jutten C. Multimodal data fusion: an overview of methods, challenges, and prospects[J]. Proceedings of the IEEE, 2015, 103(9): 1449-1477.