Cherry defect detection and recognition based on machine vision

Authors

PEI Yuekun, Beidou High Precision Positioning Service Technology Engineering Laboratory of Liaoning Province, Dalian University, Dalian, Liaoning 116622, China; Environment Sensing and Intelligent Control Key Laboratory of Dalian, Dalian University, Dalian, Liaoning
LIAN Mingyue, Beidou High Precision Positioning Service Technology Engineering Laboratory of Liaoning Province, Dalian University, Dalian, Liaoning 116622, China; Environment Sensing and Intelligent Control Key Laboratory of Dalian, Dalian University, Dalian, Liaoning
JIANG Yanchao, Beidou High Precision Positioning Service Technology Engineering Laboratory of Liaoning Province, Dalian University, Dalian, Liaoning 116622, China; Environment Sensing and Intelligent Control Key Laboratory of Dalian, Dalian University, Dalian, Liaoning
YE Jiamin, Beidou High Precision Positioning Service Technology Engineering Laboratory of Liaoning Province, Dalian University, Dalian, Liaoning 116622, China; Environment Sensing and Intelligent Control Key Laboratory of Dalian, Dalian University, Dalian, Liaoning
HAN Xinxin, Beidou High Precision Positioning Service Technology Engineering Laboratory of Liaoning Province, Dalian University, Dalian, Liaoning 116622, China; Environment Sensing and Intelligent Control Key Laboratory of Dalian, Dalian University, Dalian, Liaoning
GU Yu, Beidou High Precision Positioning Service Technology Engineering Laboratory of Liaoning Province, Dalian University, Dalian, Liaoning 116622, China; Environment Sensing and Intelligent Control Key Laboratory of Dalian, Dalian University, Dalian, Liaoning

Abstract

Based on the machine vision technology, convolutional neural network (CNN) was used to detect and recognize, and verified the cherry defects. The results showed that the recognition accuracy of intact cherry was 99.25%, with the average recognition accuracy of defective cherry of 97.99%, and the recognition speed was 25 per second. Compared with other research methods, this method could accurately detect and identify various types of defects.

Publication Date

12-28-2019

First Page

137

Last Page

140,226

DOI

10.13652/j.issn.1003-5788.2019.12.025

Recommended Citation

Yuekun, PEI; Mingyue, LIAN; Yanchao, JIANG; Jiamin, YE; Xinxin, HAN; and Yu, GU (2019) "Cherry defect detection and recognition based on machine vision," Food and Machinery: Vol. 35: Iss. 12, Article 25.
DOI: 10.13652/j.issn.1003-5788.2019.12.025
Available at: https://www.ifoodmm.cn/journal/vol35/iss12/25

References

[1] 崔建潮, 王文辉, 贾晓辉, 等. 从国内外甜樱桃生产现状看国内甜樱桃产业存在的问题及发展对策[J]. 果树学报, 2017, 34(5): 620-631.
[2] 李玉生, 程和禾, 陈龙, 等. 中国樱桃与甜樱桃种质资源在我国的分布[J]. 河北果树, 2019(2): 3-4, 7.
[3] BALESTANI A M, MOGHADDAM P A, MOTLAQ A M, et al. Sorting and grading of cherries on the basis of ripeness, size and defects by using image processing techniques[J]. International Journal of Agriculture and Crop Sciences, 2012, 4(16): 1 144-1 149.
[4] 王昭, 程朋乐, 皇甫宜龙, 等. 基于NI Vision Assistant的樱桃缺陷检测方法[J]. 计算机科学与应用, 2017, 7(12): 1 163-1 173.
[5] 吕红. 基于卷积神经网络的手写数字识别系统的设计[J]. 智能计算机与应用, 2019, 9(2): 54-56, 62.
[6] 刘建国, 代芳, 詹涛. 基于卷积神经网络的车牌识别技术[J]. 物流技术, 2018, 37(10): 62-66, 126.
[7] 刘云, 杨建滨, 王传旭. 基于卷积神经网络的苹果缺陷检测算法[J]. 电子测量技术, 2017, 40(3): 108-112.
[8] 吴志洋, 卓勇, 李军, 等. 基于卷积神经网络的单色布匹瑕疵快速检测算法[J]. 计算机辅助设计与图形学学报, 2018, 30(12): 2 262-2 270.
[9] 邡鑫, 史峥. 基于卷积神经网络的晶圆缺陷检测与分类算法[J]. 计算机工程, 2018, 44(8): 218-223.
[10] 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): 1 378-1 385.
[11] 卞国龙, 李勇, 戚顺青, 等. 基于卷积神经网络的轮胎X射线图像缺陷检测[J]. 轮胎工业, 2019, 39(4): 247-251.
[12] SRIVASTAVA N, HINTON G, KRIZHEVSKY A, et al. Dropout: A simple way to prevent neural networks from overfitting[J]. The Journal of Machine Learning Research, 2014, 15(1): 1 929-1 958.