Underwater Dense Targets Detection and Classification based on YOLOv3

In order to meet the requirements of fast detection and classification of underwater targets during intelligent underwater robot operation, an improved YOLOv3 algorithm named YOLOv3-UW algorithm is proposed to improve the detection accuracy and detection speed. Compared with the YOLOv3 algorithm, the YOLOv3-UW algorithm improves the clusters algorithm of data sets, optimizes the network structure, and improves the residual module. The final experimental results show that detection speed and detection accuracy of the YOLOv3-UW algorithm are higher than the YOLOv3 algorithm.

[1]  Da Peng Jiang,et al.  The Basic Control System of an Ocean Exploration AUV , 2013 .

[2]  Ali Farhadi,et al.  YOLO9000: Better, Faster, Stronger , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  M. H. Supriya,et al.  Deep learning architectures for underwater target recognition , 2013, 2013 Ocean Electronics (SYMPOL).

[4]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Gang Sun,et al.  Squeeze-and-Excitation Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[6]  Kevin D. LePage,et al.  An Autonomous Underwater Vehicle Data-Driven Control Strategy for Target Tracking , 2018, IEEE Journal of Oceanic Engineering.

[7]  Albert S. Huang,et al.  ORCA-IX : An Autonomous Underwater Vehicle , 2006 .

[8]  Mohammad Reza Daliri,et al.  Underwater cable detection in the images using edge classification based on texture information , 2016 .

[9]  K. D. Moore,et al.  Underwater Optical Imaging: Status and Prospects , 2001 .

[10]  Christof Koch,et al.  Detection and tracking of objects in underwater video , 2004, CVPR 2004.

[11]  David A. Clausi,et al.  K-means Iterative Fisher (KIF) unsupervised clustering algorithm applied to image texture segmentation , 2002, Pattern Recognit..

[12]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[13]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.

[14]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.