水下声呐图像轻量级目标检测模型

摘要/Abstract

摘要： 水下AUV搭载声呐进行探测成为水下目标检测的主流方式，水下环境的复杂及声呐成像方式导致声呐图像分辨率较低，使用形态学目标检测等传统方法时检测精度与实时性不高，深度学习如YOLO等算法直接用于水下声呐图像目标检测时仍然面临样本少、模型参数多等挑战，为此，本文提出一种声呐图像水下目标轻量化检测模型。针对低分辨率声呐图像数据特点以及水下AUV自动检测对实时性的要求，以YOLOv4模型为主要框架，进行模型裁剪、替换优化特征融合模块、目标预测框K均值聚类以及改进损失函数等，将构建的检测模型应用于声呐目标检测。所构建的声呐图像水下目标检测轻量化模型的mAP相对于SSD、YOLOv3、YOLOv3-DFPIN、YOLOv4-tiny分别提高了0.0659、0.0214、.0402和0.1701。在mAP相较于YOLOv4、CenterNet、EfficientdetD0分别低0.0186、0.0093、0.0074的情况下，FPS分别相对于YOLOv4提升一倍多、相对于EfficientdetD0提升近5倍、相对于CenterNet提升近一倍。同时，本文提出的模型兼具高精度和实时性的优点。实验结果表明，本文提出的特征提取网络能够减小网络参数冗余，提高模型效率和检测速度，结合自适应空间特征融合模块增强了不同尺度之间特征的相互融合和重用，提高了低分辨率声呐图像目标检测的精度。

关键词: 目标检测, 水下声呐图像, 深度学习, YOLOv4, Kmeans++

Abstract: With the development of unmanned underwater detection technology， AUV with sonar detection has become the main method of underwater object detection. However， due to the complexity of the underwater environment and the limitation of the sonar imaging mode， the sonar image resolution is low. Therefore， the traditional morphology based on object detection method has the problems of low detection accuracy and poor real-time performance. When deep learning algorithms such as YOLO are directly applied to underwater sonar image target detection， they still face challenges such as few underwater samples and many model parameters. This paper proposes a lightweight object detection model for sonar image datasets. In view of the characteristics of low-resolution sonar image data and the real-time requirements of underwater AUV automatic detection， the YOLOv4 model is used as the main framework to carry out model tailoring， replace the optimized feature fusion module， target prediction K-means clustering and improve the loss function， etc.， and the constructed detection model is applied to sonar target detection. According to the experimental data， the mAP of the proposed model in this paper is 0.0659，0.0214，0.0402 and 0.1701 higher than that of SSD， YOLOv3， YOLOV3-DFPIN and YOLOV4-tiny respectively， Under the conditioms of the mAPs are only 0.0186 lower than that of YOLOv4， only 0.0093 lower than CenterNet， only 0.0074 lower than EfficientdetD0， however， FPS is more than twice as high as YOLOv4 and CenterNet， more than fifth as high as EfficientdetD0. At the same time， the proposed model in this paper has the advantages of both high precision and real time. The experimental results show that the proposed feature extraction network can greatly reduce the redundancy of network parameters and improve the model efficiency and detection speed. Combined with the adaptive spatial feature fusion module， the mutual fusion and reuse of features in different scales are enhanced， and the accuracy of low resolution sonar image target detection is improved.

Key words: object detection, underwater sonar image, deep learning, YOLOv4, Kmeans++

范新南, 陈馨洋, 史朋飞, 孙奂儒, 鹿亮, 周仲凯. 水下声呐图像轻量级目标检测模型[J]. 计算机与现代化, 2023, 0(03): 16-22.

FAN Xin-nan, CHEN Xin-yang, SHI Peng-fei, SUN Huan-ru, LU Liang, ZHOU Zhong-kai. Lightweight Object Detection Model for Underwater Sonar Images[J]. Computer and Modernization, 2023, 0(03): 16-22.

参考文献［38］

［1］	CHOI Y W， CHUNG Y S， LEE S I， et al. Rear object detection method based on optical flow and vehicle information for moving vehicle［C］// 2017 9th International Conference on Ubiquitous and Future Networks（ICUFN）. 2017：203-205.
［2］	KANG Y K， HUANG W C， ZHENG S. An improved frame difference method for moving target detection［C］// 2017 Chinese Automation Congress（CAC）. 2017：1537-1541.
［3］	PRABOWO M R， HUDAYANI N， PURWIYANTI S， et al. A moving objects detection in underwater video using subtraction of the background model［C］// 2017 International Conference on Electrical Engineering， Computer Science and Informatics （EECSI）. 2017：1-4.
［4］	REN S Q， HE K M， GIRSHICK R， et al. Faster R-CNN： Towards real-time object detection with region proposal networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2015（28）：1137-1149.
［5］	LIN T Y， DOLLAR P， GIRSHICK R， et al. Feature pyramid networks for object detection［C］// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. 2017：936-944.
［6］	WU S， XU Y， ZHAO D N. Survey of target detection based on deep convolution network［J］. Pattern recognition and artificial intelligence， 2018，31（4）：335-346.
［7］	XU Y J， QU W Y， LI Z Y， et al. Efficient k-Means++ approximation with mapreduce［J］. IEEE Transactions on Parallel and Distributed Systems， 2014，25（12）：3135-3144.
［8］	DEWI C， CHEN R C， LIU Y T， et al. YOLOv4 for advanced traffic sign recognition with synthetic training data generated by various GAN［J］. IEEE Access， 2021（9）：97228-97242.
［9］	REDMON J， DIVVALA S， GIRSHICK R， et al. You only look once： Unified， real-time object detection［C］// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. 2016：779-788.
［10］	ZHAO L Q， LI S Y. Object detection algorithm based on improved YOLOv3［J］. Electronics， 2020，9（3）. DOI：10.3390/electronics9030537.
［11］	TIAN Y N， YANG G D， WANG Z， et al. Apple detection during different growth stages in orchards using the improved YOLOv3 model［J］. Computers and Electronics in Agriculture， 2019，157（）：417-426.
［12］	LIU T， PANG B， ZHANG L， et al. Sea surface object detection algorithm based on YOLOv4 fused with reverse depthwise separable convolution for USV［J］. Journal of Marine Science and Engineering， 2021，9（7）：753.
［13］	HE K M， ZHANG X Y， REN S Q， et al. Deep residual learning for image recognition［C］// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. 2016：770-778.
［14］	HE K M， ZHANG X Y， REN S Q， et al. Spatial pyramid pooling in deep convolutional networks for visual recognition［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2015，37（9）：1904-1916.
［15］	LIU S， QI L， QIN H F， et al. Path aggregation network for instance segmentation［C］// Proceedings of 2018 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. 2018：8759-8768.
［16］	SCHERER D， MÜLLER A， BEHNKE S. Evaluation of pooling operations in convolutional architectures for object recognition［C］// International Conference on Artificial Neural Networks （IWANN）. 2010：92-101.
［17］	BONACCORSO G. Machine Learning Algorithms［M］. Birmingham： Packt Publishing Ltd， 2017：5-8.
［18］	郭海涛，徐雷，赵红叶，等. 一种抑制声呐图像散斑噪声的形态学滤波器［J］. 仪器仪表学报， 2015，36（3）：654-660.
［19］	杨嘉诚，黄佳慧，韩永麟，等. 优化YOLOv4算法的安检X光图像检测网络［J］. 计算机系统应用， 2021，30（12）：116-122.
［20］	LIANG Y N， CHEN Z. An approach to reliability evaluation of Web services composition based on BPEL［J］. Applied Mechanics and Materials， 2011，135：198-204.
［21］	LIU L， WANG O Y， WANG X G， et al. Deep learning for generic object detection： A survey［J］. International Journal of Computer Vision， 2019，128（2）：261-318.
［22］	ALBAWI S， MOHAMMED T A， AL-ZAWI S. Understanding of a convolutional neural network［C］// 2017 International Conference on Engineering and Technology （ICET）. 2017：1-6.
［23］	SARIKAYA R， HINTON G E， DEORAS A. Application of deep belief networks for natural language understanding［J］. IEEE/ACM Transactions on Audio， Speech， and Language Processing， 2014，22（4）：778-784.
［24］	LIU K， LIU W， MA H D， et al. A real-time action representation with temporal encoding and deep compression［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2020，31（2）：647-660.
［25］	AKYOL K. Comparing of deep neural networks and extreme learning machines based on growing and pruning approach［J］. Expert Systems with Applications， 2019，140（C）. DOI：10.1016/j.eswa.2019.112875.
［26］	MOONS B， BRABANDERE B D， GOOL V L， et al. Energy-efficient ConvNets through approximate computing［C］// 2016 IEEE Winter Conference on Applications of Computer Vision （WACV）. 2016：1-8.
［27］	GU J X， WANG Z H， KUEN J， et al. Recent advances in convolutional neural networks［J］. Pattern Recognition， 2018，77（C）：354-377.
［28］	SIM H， LEE J. Log-quantized stochastic computing for memory and computation efficient DNNs［C］// Proceedings of the 24th Asia and South Pacific Design Automation Conference （ASPDAC）. 2019：280-285.
［29］	DONG S， MA Y H， LI C M. Implementation of detection system of grassland degradation indicator grass species based on YOLOv3-SPP algorithm［J］. Journal of Physics： Conference Series， 2021，1738（1）. DOI：10.1088/1742-6596/1738/1/012051.
［30］	QIU H， MA Y C， LI Z M， et al. BorderDet： Border feature for dense object detection［C］// The 16th European Conference on Computer Vision （ECCV）. 2020：549-564.
［31］	BOCHKOVSKIY A， WANG C Y， LIAO H Y M. YOLOv4： Optimal speed and accuracy of object detection［J］. arXiv preprint arXiv：2004.10934， 2020.
［32］	吴帅，徐勇，赵东宁. 基于深度卷积网络的目标检测综述［J］. 模式识别与人工智能， 2018，31（4）：335-346.
［33］	ZHAO C D， LEI Y. Intra-class cutmix for unbalanced data augmentation［C］// 2021 13th International Conference on Machine Learning and Computing （ICMLC）. 2021：246-251.
［34］	LIU W， ANGUELOV D， ERHAN D， et al. SSD： Single shot multibox detector［C］// European Conference on Computer Vision （ECCV）. 2016：21-37.
［35］	DUAN K W， BAI S， XIE L X， et al. CenterNet： Keypoint triplets for object detection［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision （ICCV）. 2019：6569-6578.
［36］	KONG W Z， HONG J C， JIA M Y， et al. YOLOv3-DPFIN： A dual-path feature fusion neural network for robust real-time sonar target detection［J］. IEEE Sensors Journal， 2019，20（7）：3745-3756.
［37］	KARACA A C. Robust and fast ship detection in SAR images with complex backgrounds based on EfficientDet model［C］// 2021 5th International Symposium on Multidisciplinary Studies and Innovative Technologies （ISMSIT）. 2021：334-339.
［38］	HOCHREITER S. The vanishing gradient problem during learning recurrent neural nets and problem solutions［J］. International Journal of Uncertainty（Fuzziness and Knowledge
	-Based Systems）， 1998，6（2）：107-116.

[1]	王秋忆, 周浩, 郑婷婷. 改进RetinaNet的电力设备目标检测方法[J]. 计算机与现代化, 2024, 0(01): 47-52.
[2]	胡崇佳, 刘金洲, 方立. 基于无监督域适应的室外点云语义分割[J]. 计算机与现代化, 2024, 0(01): 74-79.
[3]	林威. 基于自监督学习和数据回放的新闻推荐模型增量学习方法[J]. 计算机与现代化, 2023, 0(12): 1-6.
[4]	梁天恺, 黄康华, 刘凯航, 兰岚, 曾碧. 基于双向同态加密的深度联邦图片分类方法[J]. 计算机与现代化, 2023, 0(12): 36-40.
[5]	邱凯星, 冯广. 基于双重特征注意力的多标签图像分类模型[J]. 计算机与现代化, 2023, 0(12): 41-47.
[6]	张伯泉, 麦海鹏, 陈嘉敏, 逄锦聚. 基于高灰度值注意力机制的脑白质高信号分割[J]. 计算机与现代化, 2023, 0(12): 67-75.
[7]	马泽宇, 叶宁, 徐康, 王甦, 王汝传, . 基于FMCW雷达和ResNeSt-GRU的行为识别方法[J]. 计算机与现代化, 2023, 0(11): 101-107.
[8]	欧嘉城, 曾安, 金亮. 基于CP-YOLOX的冷冻电镜图像蛋白质目标检测算法[J]. 计算机与现代化, 2023, 0(11): 113-119.
[9]	闫子贤, 董宝良, 唐思谜. 针对复杂背景下低分辨率舰船目标的改进YOLOv7算法[J]. 计算机与现代化, 2023, 0(11): 120-126.
[10]	李延满, 王必恒, 赵羚焱. 基于轻量化YOLOv5的安全帽检测[J]. 计算机与现代化, 2023, 0(10): 59-64.
[11]	黎世达, 项剑文. 一种提高图像识别模型鲁棒性的弱化强化方法[J]. 计算机与现代化, 2023, 0(10): 70-76.
[12]	沈加炜, 陆一鸣, 陈晓艺, 钱美玲, 陆卫忠, . 基于深度学习的人体行为检测方法研究综述[J]. 计算机与现代化, 2023, 0(09): 1-9.
[13]	顾成伟, 丁勇, 李登华. 基于计算机视觉的工业厂区人员安全警戒系统[J]. 计算机与现代化, 2023, 0(09): 20-26.
[14]	刘禅奕, 黄丹, 薛林雁, 王涛, 朱桃, . 改进EfficientNet网络的COVID-19 X光分类[J]. 计算机与现代化, 2023, 0(09): 94-99.
[15]	马国祥, 杨凌菲, 严传波, 张志豪, 孙彬, 王晓荣. 基于深度DenseNet网络的肝包虫病超声影像诊断方法[J]. 计算机与现代化, 2023, 0(09): 100-104.