[1] [1] SAHOO ADWIVEDY S KROBI P S.Advancements in the field of autonomous underwater vehicle［J］.Ocean Engineering2019181:145160.

[2] [2] FORSYTH D.Object detection with discriminatively trained partbased models［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence201447(2):67.

[3] [3] KRIZHEVSKY ASUTSKEVER IHINTON G E.ImageNet classification with deep convolutional neural networks［J］.Communications of the ACM201760(6)：8490.

[4] [4] ZHAO Z QZHENG PXU S Tet al.Object detection with deep learning［J］.IEEE Transactions on Neural Networks and Learning Systems201930(11):32123232.

[5] [5] UIJLINGS J R RVAN DE SANDE K E AGEVERS Tet al.Selective search for object recognition［J］.International Journal of Computer Vision2013104(2)：154171.

[6] [6] GIRSHICK R.Fast RCNN［R］.Los Alamos:arXiv Preprint2015:arXiv:1504.08083.

[7] [7] REN S QHE K MGIRSHICK Ret al.Faster RCNN:towards realtime object detection with region proposalnetworks［R］.Los Alamos:arXiv Preprint2016:arXiv:1506.01497.

[8] [8] HE K MGKIOXARI GDOLLR Pet al.Mask RCNN［R］.Los Alamos:arXiv Preprint2018:arXiv:1703.06870.

[9] [9] CAI Z WVASCONCELOS N.Cascade RCNN:delving into high quality object detection［R］.Los Alamos:arXiv Preprint2018:arXiv:1712.00726.

[10] [10] LIU WANGUELOV DERHAN Det al.SSD:single shot multibox detector［R］.Los Alamos:arXiv Preprint2016:arXiv:1512.02325.

[11] [11] REDMON JDIVVALA SGIRSHICK Ret al.You only look once:unifiedrealtime object detection［R］.Los Alamos:arXiv Preprint2016:arXiv:1506.02640.

[12] [12] REDMON JFARHADI A.YOLO9000:betterfasterstronger［R］.Los Alamos:arXiv Preprint2016:arXiv:1612.08242.

[13] [13] REDMON JFARHADI A.YOLOv3:an incremental improvement［R］.Los Alamos:arXiv Preprint2018:arXiv:1804.02767.

[14] [14] BOCHKOVSKIY AWANG C YLIAO H Y M.YOLOv4:optimal speed and accuracy of object detection［R］.Los Alamos:arXiv Preprint2020:arXiv:2004.10934.

[15] [15] LI C YLI L LJIANG H Let al.YOLOv6:a singlestage object detection framework for industrial applications［R］.Los Alamos:arXiv Preprint2022:arXiv:2209.02976.

[16] [16] WANG C YBOCHKOVSKIY ALIAO H Y M.YOLOv7:trainable bagoffreebies sets new stateoftheart for realtime object detectors［R］.Los Alamos:arXiv Preprint 2022:arXiv:2207.02696.

[18] [18] CHEN XYUAN M J HYANG Qet al.UnderwaterYCC:underwater target detection optimization algorithm based on YOLOv7［J］.Journal of Marine Science and Engineering202311(5)：995.

[19] [19] WOO SPARK JLEE JYet al.CBAM:convolutional block attention module［R］.Los Alamos:arXiv Preprint 2018:arXiv:1807.06521.

[20] [20] LIN TYDOLLR PGIRSHICK Ret al.Feature pyramid networks for object detection［C］//IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Honolulu:IEEE2017.doi:10.1109/CVPR.2017.106.

[21] [21] LIU SQI LQIN H Fet al.Path aggregation network for instance segmentation［C］//IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).Salt Lake City:IEEE2018.doi:10.1109/CVPR.2018.00913.

[22] [22] MA N NZHANG X YZHENG H Tet al.ShuffleNet v2:practical guidelines for efficient CNN architecture design［R］.Los Alamos:arXiv Preprint2018:arXiv:1807.11164.

[23] [23] CHEN J RKAO SHHE Het al.Rundon’t walk:chasing higher FLOPS for faster neural networks［R］.Los Alamos:arXiv Preprint2023:arXiv:2303.03667.

[24] [24] HU JSHEN LALBANIE Set al.Squeezeandexcitation networks［R］.Los Alamos:arXiv Preprint2019:arXiv:1709.01507.

[25] [25] ZHENG Z HWANG PLIU Wet al.DistanceIoU loss:faster and better learning for bounding box regression［R］.Los Alamos:arXiv Preprint2019:arXiv:1911.08287.

[26] [26] GEVORGYAN Z.SIoU loss:more powerful learning for bounding box regression［R］.Los Alamos:arXiv Preprint2022:arXiv:2205.12740.