基于机器视觉技术的棉籽破损检测

doi:10.13733/j.jcam.issn.20955553.2022.05.011

摘要/Abstract

摘要： 种子质量对增产丰收具有十分重要的意义。提出一种基于机器视觉提取单粒棉籽，自动识别破损棉籽的方法。试验用最大类间法自动选择阈值结合膨胀处理和区域属性度量函数提取单粒棉籽图像；基于改进阈值的小波去噪对图像进行增强；通过对多幅单粒棉籽图像的研究找到对破损区域进行阈值分割的最佳阈值，对整个棉籽进行阈值分割的最佳阈值；进而对分割后的图像进行相乘和数学形态学处理等方法得到破损区域特征；最后利用获取连通区域的方法实现破损棉籽的识别并将此方法用Matlab App Designer设计成软件。试验表明，此系统对破损棉籽的平均准确率达到89%。优于软阈值函数、硬阈值函数和软硬阈值折衷函数的平均准确率83.5%、85%和87.5%。

关键词: 脱绒棉籽, 机器视觉, 小波去噪, 破损检测

Abstract: Seed quality is of great significance for increasing yield and harvest. Therefore, this paper proposes a method of extracting single cotton seed and automatically identifying damaged cotton seed based on machine vision.In this study, the maximum class method was used to automatically select the threshold value. The expansion processing and the region attribute measurement function was also used to extract the single cottonseed image. Wavelet denoising based on improved threshold was used to enhance the image. Through the study of numerous singlegrain cottonseed images, the optimal threshold for segmenting the damaged and the whole cottonseed was found. The damaged area features were obtained by multiplying the segmented images and morphological processing. Finally, the damaged cottonseeds were identified using the method of obtaining the connected region designed into a software with Matlab App Designer.The average accuracy of this system to damaged cottonseed was 89%, which was better than soft threshold function, hard threshold function and softhard threshold compromise function with average accuracy of 83.5%, 85% and 87.5%, respectively.

Key words: cashmere cottonseed, machine vision, wavelet denoising, damage detection

中图分类号:

S561： TP391

刘媛杰, 张泽亮, 张洪洲, 李勇, 李伟强. 基于机器视觉技术的棉籽破损检测[J]. 中国农机化学报, 2022, 43(5): 71-76.

Liu Yuanjie, Zhang Zeliang, Zhang Hongzhou, Li Yong, Li Weiqiang.. Detection of cottonseed damage based on machine vision technology[J]. Journal of Chinese Agricultural Mechanization, 2022, 43(5): 71-76.

参考文献

［1］王新宇, 牛鹏帅, 卢伟, 等. 基于光声光谱深度扫描的单粒玉米种子发芽率无损检测方法［J］. 华南农业大学学报, 2020,41(6): 119-125.
Wang Xinyu, Niu Pengshuai, Lu Wei, et al. A nondestructive detection method for single maize seed germination rate based on photoacoustic spectrum deep scanning ［J］. Journal of South China Agricultural University, 2020, 41(6):119-125.
［2］孙六莲. 我国色选机发展概况与建议［J］. 农业与技术, 2016, 36(9): 62-64.
［3］ Gunasekaran S, Cooper T M, Berlage A G, et al. Image processing for stress cracks in corn kernels ［J］. Transactions of the Asae, 1987, 30(1): 266-273.
［4］ Yie T J, Liao K, Paulsen M R, et al.Corn kernel stress crack detection by machine vision ［C］. American Society of Agricultural Engineers, 1993.
［5］余淑华, 刘艳丽, 王世璞, 等. 脱绒棉种色选机双CCD检测系统设计与仿真分析［J］. 农业机械学报, 2015, 46(9): 55-61.
Yu Shuhua, Liu Yanli, Wang Shipu, et al. Design and simulation analysis of dual CCD detection system for delinted cottonseeds color sorter ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(9): 55-61.
［6］张俊雄, 陈涛, 于振东, 等. 基于计算机视觉的新疆棉种颜色分选系统设计［J］. 农业机械学报, 2009, 40(10): 161-164.
Zhang Junxiong, Chen Tao, Yu Zhendong, et al. Xinjiang cotton seed color separation system based on computer vision ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2009, 40(10): 161-164.
［7］张洪洲. 棉籽分级系统设计与PLC实现［J］. 中国农机化学报, 2015, 36(2): 113-115.
Zhang Hongzhou. Design and PLC implementation of the classifying system for cottonseeds ［J］. Journal of Chinese Agricultural Mechanization, 2015, 36(2): 113-115.
［8］张新伟, 赵学观, 张健东, 等. 基于数据融合的玉米种子内部机械裂纹检测方法［J］. 农业工程学报, 2012, 28(9): 136-141.
Zhang Xinwei, Zhao Xueguan, Zhang Jiandong, et al. Detection of internal mechanical cracks in corn seeds based on data fusion technology ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(9): 136-141.
［9］刘付桂兰. 基于MATLAB图像处理技术及应用［J］. 科学技术创新, 2020(35): 64-67.
［10］赵玉水, 张聿, 王蕾. Mandelbrot集图形的数学形态学处理方法与丝织应用［J］. 丝绸, 2011, 48(6): 22-24.
Zhao Yushui, Zhang Yu, Wang Lei, et al. Method of processing Mandelbrot set graphics and its application for silk jacquard fabric pattern based on mathematical morphology ［J］. Silk Monthly, 2011, 48(6): 22-24.
［11］ Wang Q P , Wang T P , Zhang K . Image edge detection based on the grey prediction model and discrete wavelet transform ［J］. Kybernetes, 2011, 41(5-6).
［12］ Kobylin O, Lyashenko V.Comparison of standard image edge detection techniques and of method based on wavelet transform ［J］. International Journal of Advanced Research, 2014, 2(8): 572-580
［13］杨章静, 钱建军, 黄璞, 等. 基于Gabor变换的GrabCut纹理图像分割［J］. 控制与决策, 2016, 31(1): 149-154.
Yang Zhangjing, Qian Jianjun, Huang Pu, et al. GrabCut texture image segmentation based on Gabor wavelet transform ［J］. Control and Decision, 2016, 31(1): 149-154.
［14］蔡乐黎. 安卓平台下基于改进的OneCut算法的目标抠取与应用研究［D］. 长沙: 湖南师范大学, 2017.
Cai Leli. Research on target extraction and application based on improved OneCut algorithm on the Android platform ［D］. Changsha: Hunan Normal University, 2017.
［15］胡薇薇. 基于Grab Cut的快速交互式图像分割算法研究［D］. 福州: 福州大学, 2013.
Hu Weiwei. Reserch of fast interactive image segmentation algorithm based on Grab Cut ［D］. Fuzhou: Fuzhou University, 2013.
［16］张绘娟, 张达敏, 闫威, 等. 基于改进阈值函数的小波变换图像去噪算法［J］. 计算机应用研究, 2020, 37(5): 1545-1548.
Zhang Huijuan, Zhang Damin, Yan Wei, et al. Wavelet transform image denoising algorithm based on improved threshold function ［J］. Application Research of Computers, 2020, 37(5): 1545-1548.
［17］张师语, 刘章玉, 石仁翠. 基于改进阈值和分层阈值的小波变换图像去噪［J］. 现代计算机, 2020(32): 52-58.
Zhang Shiyu, Liu Zhangyu, Shi Rencui. Wavelet transformation image denoising based on improved threshold and hierarchical threshold ［J］. Modern Computer, 2020(32): 52-58.

[1]	王珊, 薛新宇, 郭祥雨. 基于机器视觉的油菜菌核病分级检测研究[J]. 中国农机化学报, 2022, 43(5): 47-53.
[2]	郑如新, 孙青云, 肖国栋. 基于机器视觉的金银花图像识别处理算法研究[J]. 中国农机化学报, 2022, 43(4): 153-159.
[3]	冯凯, 王茂励, 何晓宁, 王东伟, 岳丹松, 初香港. 基于辅助线的农机视觉导航路径提取算法研究[J]. 中国农机化学报, 2022, 43(10): 167-175.
[4]	李昕, 陈泽君, 李立君, 谭季秋, 吴发展, . 一种基于偏好免疫网络多特征辨识的油茶果分选识别方法[J]. 中国农机化学报, 2021, 42(9): 187-194.
[5]	白云龙;傅彬;史振华;王健;. 基于标记分水岭算法的插秧机器人导航路径检测[J]. 中国农机化学报, 2021, 42(7): 142-147.
[6]	刘海涛;伊丽丽;兰玉彬;韩鑫;崔立华;. 机器视觉在棉花智能打顶领域的应用研究进展[J]. 中国农机化学报, 2021, 42(6): 159-165.
[7]	吕沐华;丁珠玉;. 基于机器视觉的果园喷药除草机器人视觉系统设计[J]. 中国农机化学报, 2021, 42(5): 42-48.
[8]	郭祥雨;薛新宇;. 基于机器视觉的水稻制种田导航线提取方法[J]. 中国农机化学报, 2021, 42(5): 197-201.
[9]	吕梦棋;张芮祥;贾浩;马丽;. 基于改进ResNet玉米种子分类方法研究[J]. 中国农机化学报, 2021, 42(4): 92-98.
[10]	杨涛;李晓晓;. 机器视觉技术在现代农业生产中的研究进展[J]. 中国农机化学报, 2021, 42(3): 171-181.
[11]	谢仕烁;黄伟锋;朱立学;杨尘宇;张世昂;付根平;. 基于改进Floodfill算法的田垄视觉导航系统[J]. 中国农机化学报, 2021, 42(3): 182-188.
[12]	金耀;何秀文;万世主;刘仁鑫;. 基于YOLO v3的生猪个体识别方法[J]. 中国农机化学报, 2021, 42(2): 178-183.
[13]	张文蓉, 王朋, 钟鸣, 黄博, 曹景军, 柴秀娟, . 基于视觉的茭白自动分级包装设备研究与开发[J]. 中国农机化学报, 2021, 42(12): 114-121.
[14]	孙进;张洋;王宁;张道周;丁煜;习俊通;. 融合机器视觉和CAN总线的玉米种粒分类器设计与试验[J]. 中国农机化学报, 2020, 41(8): 81-89+120.
[15]	卜小东;张馨月;黄可京;郭辉;. 基于机器视觉的玉米苗期土壤湿度检测试验研究[J]. 中国农机化学报, 2020, 41(7): 150-156.