[1]翟先一,魏鸿磊,韩美奇,等.基于改进YOLO卷积神经网络的水下海参检测[J].江苏农业学报,2023,(07):1543-1553.[doi:doi:10.3969/j.issn.1000-4440.2023.07.011]
　ZHAI Xian-yi,WEI Hong-lei,HAN Mei-qi,et al.Underwater sea cucumber identification based on improved YOLO convolutional neural network[J].,2023,(07):1543-1553.[doi:doi:10.3969/j.issn.1000-4440.2023.07.011]
点击复制

基于改进YOLO卷积神经网络的水下海参检测()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2023年07期
页码:
1543-1553
栏目:
农业信息工程
出版日期:
2023-10-31

文章信息/Info

Title:
Underwater sea cucumber identification based on improved YOLO convolutional neural network
作者:
翟先一1; 魏鸿磊1; 韩美奇2; 黄萌1
(1.大连工业大学机械工程与自动化学院，辽宁大连116034；2.中国科学院空天信息创新研究院/传感技术国家重点实验室，北京100190)
Author(s):
ZHAI Xian-yi1; WEI Hong-lei1; HAN Mei-qi2; HUANG Meng1
(1.School of Mechanical Engineering and Automation, Dalian Polytechnic University，Dalian 116034, China；2.Aerospace Information Research Institute， Chinese Academy of Sciences/State Key Laboratory of Transducer Technology, Beijing 100190, China)
关键词:
YOLO; 目标检测; 深度学习; 机器视觉; 卷积神经网络
Keywords:
YOLO (You only look once); object identification; deep learning; computer vision; convolutional neural network
分类号:
S126
DOI:
doi:10.3969/j.issn.1000-4440.2023.07.011
文献标志码:
A
摘要:
为了实现水下海参的自动化捕捞，需要利用机器视觉方法实现水下海参的实时检测与定位。本研究提出一种基于改进YOLOv5s的水下海参检测定位方法。针对海参与水下环境对比度较低的问题，引入多尺度视觉恢复算法对图像进行处理，增强图像对比度；为了提高模型特征提取能力，加入了注意力机制模块；原始模型对YOLOv5s小目标的检测效果不佳，改进后的YOLOv5s模型替换了原有的激活函数，并在Head网络中加入了新的针对小目标的Detect层。使用改进的YOLOv5s模型与YOLOv5s、YOLOv4和Faster-RCNN在相同的图像数据集上进行试验，结果表明，改进的YOLOv5s模型的检测精度和置信度，尤其是对小目标的检测效果优于其他模型。与YOLOv5s模型相比，改进后的YOLOv5s模型的精度和召回率分别提高了9.6个百分点和12.4个百分点，能够满足水下海参的实时检测要求。
Abstract:
To realize automatic fishing of sea cucumber underwater, it is necessary to use machine vision method to realize real-time detection and positioning of underwater sea cucumber. In this study, a detection and localization method for underwater sea cucumber based on improved YOLOv5s was proposed. Aiming at the problem of low contrast between the sea cucumber and the underwater environment, a multi-scale vision restoration algorithm was introduced to process the images to enhance the contrast of the images. The attention mechanism module was added to improve the feature extraction ability of the model. The original model didn’t show good detection effect on small object of YOLOv5s. The improved YOLOv5s model replaced the original activation function and added a new Detect layer into the Head network which aimed at small object. The improved YOLOv5s model was used to conduct experiments with YOLOv5s, YOLOv4 and Faster-RCNN on the same image data set. The results showed that, the improved YOLOv5s model showed better detection accuracy and degree of confidence compared with other models, especially for small target detection. Compared with the YOLOv5s model, the precision and recall rate of the improved YOLOv5s model increased by 9.6 percentage points and 12.4 percentage points respectively, which could meet the requirement of real-time detection of underwater sea cucumber.

参考文献/References:

[1]刘吉伟,魏鸿磊,裴起潮,等. 采用相关滤波的水下海参目标跟踪[J]. 智能系统学报, 2019, 77(3): 525-532.
[2]李娟,朱学岩,葛凤丽,等. 基于计算机视觉的水下海参检测方法研究[J]. 中国农机化学报, 2020, 41(7): 171-177.
[3]强伟,贺昱曜,郭玉锦,等. 基于改进SSD的水下目标检测算法研究[J]. 西北工业大学学报, 2020, 38(4): 747-754.
[4]林宇,何水原. 改进Cascade RCNN的水下目标检测[J]. 电子世界, 20221(1):105-108.
[5]马孔伟. 基于CNN的海参检测技术及其在水下机器人中的应用[D]. 哈尔滨:哈尔滨工业大学, 2019: 56.
[6]王璐,王雷欧,王东辉. 基于Faster-rcnn的水下目标检测算法研究[J]. 网络新媒体技术, 2021, 10(5): 43-51，58.
[7]徐建华,豆毅庚,郑亚山. 一种基于YOLO-V3算法的水下目标检测跟踪方法[J]. 中国惯性技术学报, 2020, 28(1): 129-133.
[8]张聪辉. 基于YOLO-v3的海参目标检测系统的设计与实现[D]. 哈尔滨:哈尔滨工程大学,2020.
[9]朱世伟,杭仁龙,刘青山. 基于类加权YOLO网络的水下目标检测[J]. 南京师大学报(自然科学版), 2020, 43(1): 129-135.
[10]WANG J, LU K, XUE J, et al. Single image dehazing based on the physical model and MSRCR algorithm[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2017, 28(9): 2190-2199.
[11]LIU Y, YAN H, GAO S, et al. Criteria to evaluate the fidelity of image enhancement by MSRCR[J]. IET Image Processing, 2018, 12(6): 880-887.
[12]许伟栋,赵忠盖. 基于卷积神经网络和支持向量机算法的马铃薯表面缺陷检测[J].江苏农业学报,2018, 34(6): 1378-1385.
[13]KUMARI N, RUF V, MUKHAMETOV S, et al. J mobile eye-tracking data analysis using object detection via YOLO v4[J]. Sensors,2021,21(22): 7668.
[14]张开兴,吕高龙,贾浩,等． 基于图像处理和BP神经网络的玉米叶部病害检测[J]. 中国农机化学报, 2019, 40(8): 122-126.
[15]尚志军,张华,曾成,等. 基于机器视觉的枳壳自动定向方法与试验[J]. 中国农机化学报, 2019, 40(7): 119-124.
[16]LIU S, XU Y, GUO L M, et al. Multi-scale personnel deep feature detection algorithm based on Extended-YOLOv3[J]. Journal of Intelligent & Fuzzy Systems, 2021, 40(1): 773-786.
[17]JOBSON D J, RAHMAN Z, WOODELL G A. A multiscale retinex for bridging the gap between color images and the human observation of scenes[J]. IEEE Transactions on Image Processing，1997, 6(7): 965-976.
[18]WANG Q Z, LI S, QIN H, et al. Super-resolution of multi-observed RGB-D images based on nonlocal regression and total variation[J]. IEEE Transactions on Image Processing: A Publication of the IEEE Signal Processing Society, 2016, 25(3): 1425-1440.
[19]ZHANG M H, XU S B, SONG W, et al. Lightweight underwater object detection based on YOLOv4 and multi-scale attentional feature fusion[J]. Remote Sensing,2021,13(22): 4706.
[20]TAN L, HUANGFU T R, WU L, et al. Comparison of RetinaNet, SSD, and YOLOv3 for real-time pill identification[J]. BMC Medical Informatics and Decision Making,2021,21(1): 1-11.
[21]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.
[22]朱 磊. Retinex图像增强算法的研究与FPGA实现[D]. 北京：清华大学, 2012.
[23]朱秋旭,李俊山,朱英宏,等.Retinex理论下的自适应红外图像增强[J]. 微电子学与计算机, 2013, 30(4): 22-25.
[24]陈旭君,王承祥,朱德泉,等. 基于YOLO卷积神经网络的水稻秧苗行线检测[J].江苏农业学报,2020,36(4):930-935.
[25]LEI F, TANG F, LI S. Underwater target detection algorithm based on improved YOLOv5[J]. Journal of Marine Science and Engineering，2022, 10：310.
[26]QIAO X, BAO J, ZHANG H, et al. fvUnderwater sea cucumber identification based on principal component analysis and support vector machine[J]. Measurement，2019, 133：444-455.
[27]ZHANG L, XING B, WANG W et al. Sea cucumber detection algorithm based on deep learning[J]. Sensors，2022, 22：5717.
[28]XUE L, ZENG X, JIN A. A novel deep-learning method with channel attention mechanism for underwater target recognition [J]. Sensors，2022, 22：5492.
[29]孟亮,郭小燕,杜佳举，等. 一种轻量级CNN农作物病害图像识别模型[J].江苏农业学报,2021,37(5):1143-1150.

相似文献/References:

[1]李恒,南新元,高丙朋,等.一种基于GhostNet的绿色类圆果实识别方法[J].江苏农业学报,2023,(03):724.[doi:doi:10.3969/j.issn.1000-4440.2023.03.013]
　LI Heng,NAN Xin-yuan,GAO Bing-peng,et al.A green round-like fruits identification method based on GhostNet[J].,2023,(07):724.[doi:doi:10.3969/j.issn.1000-4440.2023.03.013]

备注/Memo

备注/Memo:
收稿日期：2022-09-14基金项目：辽宁省教育厅2021年度科学研究经费面上项目(LJKZ0535、LJKZ0526)；2021年度大连工业大学本科教育教学综合改革项目(JGLX2021020、 JCLX2021008)；大连工业大学研究生创新基金项目(2023CXYJ13)作者简介：翟先一(1998-)，男，天津人，硕士研究生，研究方向为机器视觉和目标检测。(E-mail)zhaixianyi9421@foxmail.com通讯作者：魏鸿磊，(E-mail)weihl2005@163.com
更新日期/Last Update: 2023-11-17