Please wait a minute...
吉林化工学院学报, 2022, 39(9): 80-85     https://doi.org/10.16039/j.cnki.cn22-1249.2022.09.017
  本期目录 | 过刊浏览 | 高级检索 |
基于深度学习的核桃外壳缺陷检测
余思黔,赵麒荣,林嘉晨,贾雁飞*,陈广大
北华大学 电气与信息工程学院,吉林,132013
Walnut Shell Defect Detection based on Deep Learning
YU Siqian, ZHAO Qirong,LIN Jiachen ,JIA Yanfei* ,CHEN Guangda
下载:  PDF (2374KB) 
输出:  BibTeX | EndNote (RIS)      
摘要 

实现了对核桃外壳缺陷的快速识别,提高基于机器视觉的核桃分选效率,提出了一种基于改进的  YOLOv5s核桃外壳缺陷检测方法。YOLOv5s网络中大量采用卷积核为3的卷积进行特征提取,为降低网络的计算量,本文提出利用深度可分离卷积代替残差网络中所采用的卷积核为3的卷积,提高对核桃外壳检测的速度。此外,为了保证精度能够满足要求,采用了改进的均值聚类对检测框进行初始化,提高生成的检测框的质量,进而提高核桃外壳缺陷检测精度。由于聚类方法相对整个网络结构计算量较小,因此对核桃外壳检测的速度影响较小。通过实验对比分析,改进后的YOLOv5s能够快速识别出核桃外壳缺陷,而且识别精度基本保持不变。

服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
余思黔
赵麒荣
林嘉晨
贾雁飞
陈广大
关键词:  深度学习  机器视觉  深度可分离卷积  核桃外壳缺陷检测     
Abstract: 

In order to realize the rapid recognition of walnut shell defects and improve the walnut sorting efficiency based on machine vision, a walnut shell defect detection method based on improved yolov5s is proposed. In yolov5s network, convolution with convolution kernel size 3 is widely used for feature extraction. In order to reduce the amount of network calculation, this paper proposes to use depth wise separable convolution instead of convolution with convolution kernel size 3 in residual network, so as to improve the speed of walnut shell detection. In addition, in order to ensure that the accuracy can meet the requirements, this paper also uses the improved mean clustering to initialize the detection frame, improve the quality of the generated detection frame, and then improve the detection accuracy of walnut shell defects. Because the clustering method has less computation than the whole network structure, it has little impact on the speed of peach shell detection. Through experimental comparison and analysis, the improved yolov5s can quickly identify walnut shell defects, and the recognition accuracy remains basically unchanged.

Key words:  deep Leaning    machine vision    depth wise separable convolution    walnut shell defect detection
               出版日期:  2022-09-25      发布日期:  2022-09-25      整期出版日期:  2022-09-25
TP398.1  
引用本文:    
余思黔, 赵麒荣, 林嘉晨, 贾雁飞, 陈广大. 基于深度学习的核桃外壳缺陷检测 [J]. 吉林化工学院学报, 2022, 39(9): 80-85.
YU Siqian, ZHAO Qirong, LIN Jiachen , JIA Yanfei , CHEN Guangda. Walnut Shell Defect Detection based on Deep Learning . Journal of Jilin Institute of Chemical Technology, 2022, 39(9): 80-85.
链接本文:  
https://xuebao.jlict.edu.cn/CN/10.16039/j.cnki.cn22-1249.2022.09.017  或          https://xuebao.jlict.edu.cn/CN/Y2022/V39/I9/80
[1] 张天驰, 于紫南. 基于自定义深度学习网络双馈风机接入弱电网故障识别[J]. 吉林化工学院学报, 2024, 41(5): 54-59.
[2] 王敏. 自适应学习中基于CNN和IIDLA的图像识别方法研究[J]. 吉林化工学院学报, 2024, 41(3): 56-61.
[3] 谢延楠, 刘兴德. 基于机器视觉的刹车盘工件上下料系统[J]. 吉林化工学院学报, 2024, 41(1): 76-81.
[4] 董如意, 崔冉. 基于YOLOV5L的交通信号灯识别研究 [J]. 吉林化工学院学报, 2023, 40(9): 37-42.
[5] 林建华, 林二妹. 一种智能果蔬识别称重系统的设计 [J]. 吉林化工学院学报, 2023, 40(7): 83-87.
[6] 张 鋆, 李温温. RRT algorithm; path planning; target offset; variable step size; cubic B-spline curve 基于改进YOLOv5s的玉米田间杂草检测方法 [J]. 吉林化工学院学报, 2023, 40(5): 26-33.
[7] 金何. 基于深度学习融合网络的交直流电网故障诊断方法研究 [J]. 吉林化工学院学报, 2023, 40(3): 93-98.
[8] 任洛莹 , 刘兴德 , 谢延楠 , 胡文松 , 余鹏泽, 孔志成. CGB-YOLO:用于检测钢铁表面缺陷的YOLO算法 [J]. 吉林化工学院学报, 2023, 40(11): 38-44.
[9] 辛 鹏, 杨剀勋, 文孝强. 基于改进SE-CNN的风电机组故障诊断方法研究 [J]. 吉林化工学院学报, 2023, 40(1): 34-40.
[10] 刘麒, 尹港 , 王影, 叶泽 . 基于深度学习的水面漂浮物识别算法设计 [J]. 吉林化工学院学报, 2022, 39(7): 28-33.
[11] 辛瑞昊 , 王甜甜 , 苗冯博 , 董哲原 , 马占森 , 冯欣. 基于深度学习的机械轴承故障智能诊断 [J]. 吉林化工学院学报, 2022, 39(11): 25-29.
[12] 王雪晴, 刘锦涛, 卫亚博, 赵换丽. 基于机器视觉和LabVIEW的有机玻璃缺陷检测研究 [J]. 吉林化工学院学报, 2021, 38(1): 68-73.
[13] 李百明. 基于机器视觉的钢珠直径测量系统设计 [J]. 吉林化工学院学报, 2021, 38(1): 58-62.
[14] 朱莉, 陈辉. 基于深度学习的单幅图像三维重建算法 [J]. 吉林化工学院学报, 2020, 37(1): 58-62.
[1] WANG Ya-hong, JIA Ying-chao. Optimization for Soxhlet Extraction of Total Flavonoids from Eggplant Root by Using Response Surface Methodology[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(3): 1 -6 .
[2] PAN Hong-wei, ZHOU Hong-li. Research Progress on Extraction Technology and Component Analysis of Volatile Flavor Compounds in Plants[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(3): 7 -10 .
[3] ZHOU Zhen-qiang, WANG Kai-bao, CHEN Yu-tian, HAN Ji, HOU Dai-bing, GUO Wen-qi. Analysis of the Design and Dynamic Characteristics of Recyclable Garbage Crushing Device[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(3): 31 -35 .
[4] ZHU Qi-rui, ZHANG Yu-feng, LI Yang-xin, WU Dong-jun, BI Shuai-nan, ZHOU Rui-jie, YOU Sai-sai. The Structural Design and Parameter Analysis of the Chair Structure of Quadriplegia of Cerebral Palsy[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(3): 40 -44 .
[5] ZHANG Yun-peng. Preparation and Characterization of Er3+Y3+Yb3+ Co-doped Al2O3 Luminescence Materials[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(3): 58 -63 .
[6] MI Ya-wei. Proof of the Manifold of n Manifold[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(3): 64 -66 .
[7] XU You-zhuan, ZHOU Hou-qing. The Lower Bounds of Energy for Some Circulant Graph[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(3): 67 -72 .
[8] ZHOU Hong-li, ZHANG Yang, CUI Hao, XUE Jian-fei. Primary Exploration of the Applied Pharmaceutical Talents Training in Ordinary Engineering Colleges Based on Social Needs and R ecruitment Drives[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(2): 1 -4 .
[9] YANG Xiu-dong, WANG Ya-hong, ZHANG Pin-mei, YANG Yan-jun, CUI Hao, ZHOU Hong-li. Discussion on the Practice Teaching Reform of Pharmaceutical Engineering Based on Engineering Education Professional Accreditation[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(2): 5 -7 .
[10] LI Long, WANG Xue-mei. Research on Higher Education Assessment Mode Based on Formative Evaluation and Information Entropy Theory[J]. Journal of Jilin Institute of Chemical Technology, 2018, 35(2): 8 -12 .
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed