基于机器视觉的核桃仁动态分级研究

doi:10.6048/j.issn.1001-4330.2018.05.014

摘要/Abstract

摘要： 【目的】基于机器视觉技术研究出一种适合新疆核桃仁动态分级处理的方法。【方法】利用实时采集且已经完成图像预处理的样品核桃图像得到核桃仁特征集合,运用mRMR特征选择算法筛选原始特征集并对特征的重要性进行排列,通过对支持向量机、决策树和朴素贝叶斯三种机器学习算法进行模型训练和测试,得出最佳分级方法,设计核桃仁自动追踪方法和动态分级流程,构建的核桃仁自动分级系统。【结果】在使用特征bin19、K₁和bin15训练朴素贝叶斯分类器时,核桃仁的分级正确率达到最大为97.33%,在动态条件下运用构建的核桃仁自动分级系统对150个核桃仁进行分级测试,正确率为81.33%。【结论】基于机器视觉研究出的核桃仁特征提取与分级方法、核桃仁动态分级处理动作方法,可以有效完成对核桃颜色和完整度的分级。

关键词: 核桃仁; 分级; 机器视觉; 动态

Abstract: 【Objective】 The purpose of this project is to study a kind of dynamic grading method of walnut kernel based on machine vision technology that will be suitable for Xinjiang.【Method】 Walnut kernel feature set was obtained from walnut image collected in real time and completed image preprocessing. Then mRMR feature selection algorithm was used to filter the original feature set and arrange the importance of the feature. Finally, support vector machine was used to analyze the importance of the feature. Three machine learning algorithms, decision tree and naive Bayes, were trained and tested, and the optimal classification method was obtained. Finally, the automatic tracking method and dynamic grading process of walnut kernel were designed, and the automatic classification system of walnut kernel was constructed.【Result】 When using feature bin19, K1 and bin15 to train naive Bayesian classifier, the classification accuracy rate of walnut kernel classification was 97.33%. Under the dynamic condition, the walnut kernel automatic grading system was used to classify 150 walnut kernel,and the overall accuracy rate was 81.33%.【Conclusion】 Based on the feature extraction and grading method of walnut kernel developed by machine vision, the method of walnut kernel dynamic grading can effectively complete the classification task of walnut color and integrity.

Key words: short ultivars; growth stages; major gene plus polygene inheritance; genetic analysis

中图分类号:

S226.5

周军, 郭俊先, 张静, 姜彦武, 艾力·哈斯木, 蔡建. 基于机器视觉的核桃仁动态分级研究[J]. 新疆农业科学, 2018, 55(5): 901-911.

ZHOU Jun, GUO Jun-xian,ZHANG Jing, JIANG Yan-wu, Eli Hasim, CAI Jian. Study on Dynamic Grading Method of Walnut Kernel Based on Machine Vision[J]. Xinjiang Agricultural Sciences, 2018, 55(5): 901-911.

参考文献 25

[1]	刘建军,姚立健,彭樟林.基于机器视觉的山核桃等级检测技术[J].浙江农业大学报,2010,(6):854-858.LIU Jian-jun, YAO Li-jian, PENG Zhang-lin. (2010). Detection technique for cathay hickory grade based on machine vision [J]. Acta Agriculturae Zhejiangensis , (6):854-858.(in Chinese)
[2]	Chen, L. N., Ma, Q. G., Chen, Y. K., Wang, B. Q., & Pei, D. (2014). Identification of major walnut cultivars grown in china based on nut phenotypes and SSR markers. Scientia Horticulturae, 168(168): 240-248.
[3]	方建军,张仕良,张虎.基于机器视觉的板栗实时分级系统[J].轻工机械,2014,(3):92-94.FANG Jian-jun, ZHANG Shi-liang, ZHANG Hu. (2014). Real-time Chestnut Sorting System by Machine Vision [J]. Light Industry Machinery, (3): 92-94.(in Chinese)
[4]	张仕良.基于机器视觉的板栗分级设备研究[D].北京:北方工业大学,2005:16-38.ZHANG Shi-liang. (2005). Research on classification equipment of Chinese chestnut based on machine vision [D]. Master Dissertation. North China University of Technology, Beijing. 16-38.(in Chinese)
[5]	季雪娇,尹成名,朱元波,等.板栗图像的低层处理及边缘检测[J].农产品加工(学刊),2008,(1):26-28.JI Xue-jiao, YIN Chen-ming, ZHU Yuan-bo, et al. (2008). Chestnut Image Primarily Processing and Edge Detecting [J]. Academic Periodical of Farm Products Processing, (1):26-28.(in Chinese)
[6]	韩志忠,赵友刚.基于计算机视觉的花生品质检测方法[J].中国粮油学报,2009,24(5):123-126.HAN Zhi-zhong, ZHAO You-gang. (2009). The method of detection of peanut quality based on computer vision [J]. Journal of the Chinese Cereals and Oils Association, 24(5): 123-126. (in Chinese)
[7]	展慧,李小昱,王为,等.基于神经网络和图像处理的板栗分级检测方法研究[C].中国农业工程学会.纪念中国农业工程学会成立30 周年暨中国农业工程学会2009 年学术年会(CSAE2009),2009.ZHAN Hui, LI Xiao-yu, WANG Wei, et al. (2009). Study on detection method of neural network and image processing based on grading China Institute of agricultural engineering[C]. Chinese Memorial Institute of agricultural engineering was founded in 30th anniversary and Chinese Institute of Agricultural Engineering Conference in 2009 (CSAE 2009, 2009.) (in Chinese)
[8]	刘军,郭俊先,帕提古丽·司拉木,等. 基于机器视觉与支持向量机的核桃外部缺陷判别分析方法[J].食品科学,2015,36(20):211-217.LIU Jun, GUO Jun-xian, Patiguli Slamu, (2015). DISCRIMINATion method of walnut external defects based on machine vision and support vector machine [J]. Food Science, 36(20): 211-217.(in Chinese)
[9]	周竹,黄懿,李小昱,等. 基于机器视觉的马铃薯自动分级方法[J]. 农业工程学报, 2012,28(7): 186-191.ZHOU Zhu, HUANG Yi, LI Xiao-yu, et al. (2012). Automatic detecting and grading method of potatoes based on machine vision [J]. Transactions of the Chinese Society of Agricultural Engineering, 28(7): 186-191.(in Chinese)
[11]	张塔烺,应宏微,邱雪娜,等.山核桃图像头尾检测方法[J].福建农业科技,2017,(7):36-39.ZHANG Ta-lang, YING Hong-wei, QIU Xue-na, et al. (2017). Image inspection for head and tail of Carya cathayensis [J]. Fujian Agricultural Science and Technology, (7): 36-39. (in Chinese)
	童钊,廖桂平,李锦卫,等. 机器视觉技术在农产品检测中的应用研究[J]. 农业网络信息, 2008, (11): 18-21.TONG Zhao, LIAO Gui-ping, LI Jin-wei, et al. (2008). Application of machine vision technique in testing of agricultural products [J]. Agricultural Network Information, (11): 18-21.(in Chinese)
[12]	简丽琼. 基于Hu矩和Zernike矩的文字识别[J]. 科技信息, 2009, (17): 460-461.JIAN Li-qiong. (2009). Text recognition based on Hu moment and Zernike moment [J]. Science and Technology Information, (17): 460-461. (in Chinese)
[13]	丁建睿,黄剑华,刘家锋,等. 基于mRMR和SVM的弹性图像特征选择与分类[J]. 哈尔滨工业大学学报,2012,44(5):86-90.DING Jian-rui, HUANG Jian-hua, LIU Jia-feng, et al. (2012). Selection and classification of elastic image features based on mRMR and SVM [J]. Journal of Harbin Institute of Technology, 44(5):86-90.(in Chinese)
[14]	张晶,毕佳佳,刘炉. 基于mRMR的多关系朴素贝叶斯分类[J]. 计算机应用与软件, 2016,33(8):57-61.ZHANG Jing, BI Jia-jia, LIU Lu. (2016). Multi-relationship naive Bayes classification based on mRMR [J]. Computer Application and Software, 33(8): 57-61.(in Chinese)
[15]	范雪莉,冯海泓,原猛. 基于互信息的主成分分析特征选择算法[J]. 控制与决策,2013, 28(6):117-121.FAN Xue-li, FENG Hai-hong, YUAN Meng. (2013). Selection algorithm of principal component analysis based on mutual information [J]. Control and Decision-Making, 28(6): 117-121.(in Chinese)
[16]	王峥琦. 基于决策树算法的改进与应用[D]. 西安:西安科技大学硕士论文,2005.WANG Zheng-qi. (2005). Improvement and application of decision tree algorithm [D]. Master Dissertation. Xi'an University of Science and Technology, Xi'an. (in Chinese)
[17]	杨静,张楠男,李建,等. 决策树算法的研究与应用[J]. 计算机技术与发展,2010,20(2): 114-116.YANG Jing, ZHANG Nan-nan, LI Jian, et al. (2010). Research and application of decision tree algorithm [J]. Computer Technology and Development, 20(2): 114-116.(in Chinese)
[18]	王峻. 朴素贝叶斯分类模型的研究与应用[D]. 合肥:合肥工业大学硕士论文,2006.WANG Jun. (2006). Research and application of naive Bayesian classification model[D].Master Dissertation. Hefei University of Technology, Hefei. (in Chinese)
[19]	李冬梅. 朴素贝叶斯与决策树混合分类方法的研究[D]. 大连:大连海事大学硕士论文,2016.LI Dong-mei. (2016). Study on hybrid classification method of naive Bayes and decision tree [D]. Master Dissertation. Dalian Maritime University, Dalian. (in Chinese)
[20]	阿曼. 朴素贝叶斯分类算法的研究与应用[D]. 大连:大连理工大学硕士论文,2014.Aman. (2014). Research and application of Simple Bias classification algorithm [D]. Master Diisertation. Dalian University of Technology, Dalian. (in Chinese)
[21]	丁世飞,齐丙娟,谭红艳. 支持向量机理论与算法研究综述[J]. 电子科技大学学报,2011,40(1):2-10.DING Shi-fei, QI Bing-juan, TAN Hong-yan. (2011). A review of the theory and algorithm of support vector machines [J]. Journal of the University of Electronic Science and Technology, 40(1): 2-10.(in Chinese)
[22]	郎宇宁,蔺娟如. 基于支持向量机的多分类方法研究[J]. 中国西部科技,2010,9(17):28-29.LANG Yu-ning, LIN Juan-ru. (2010). Research on support vector machine classification method [J]. China Western Science and Technology, 9(17): 28-29. (in Chinese)
[23]	梁栋,管青松,黄文江,等. 基于支持向量机回归的冬小麦叶面积指数遥感反演[J]. 农业工程学报, 2013, 29 (7):125-131.LIANG Dong, GUAN Qing-song, HUANG Wen-jiang, et al. (2013). Remote sensing inversion of winter wheat leaf area index based on support vector machine regression [J]. Journal of Agricultural Engineering, 29(7): 125-131. (in Chinese)
[24]	王露,龚光红. 基于ReliefF_mRMR特征降维算法的多特征遥感图像分类[J]. 中国体视学与图像分析,2014,19(3):34-41.WANG Lu, GONG Guang-hong. (2014). Classification of multi feature remote sensing images based on ReliefF_mRMR feature reduction algorithm [J]. Chinese Journal of Stereology and Image Analysis,19(3): 34-41. (in Chinese)
[25]	黄莉莉,汤进,孙登第,等. 基于多标签ReliefF的特征选择算法[J]. 计算机应用, 2012, 32(10): 212-214.HUANG Li-li, TANG Jin, SUN Deng-di, et al. (2012). Feature selection algorithm based on multi-label ReliefF [J]. Journal of Computer Applications, 32(10): 212-214. (in Chinese)

[1]	徐静, 石书兵, 秦小钢, 朱军. 核桃林间作西红花对其土壤微生物数量的动态变化[J]. 新疆农业科学, 2023, 60(8): 2022-2027.
[2]	冯倩, 何齐, 李雪, 易鸳鸯, 孙建, 朱静, 顾美英, 谭慧林, 张志东. 阿克苏野生牛肝菌多糖提取优化及不同组分抗氧化活性分析[J]. 新疆农业科学, 2023, 60(7): 1813-1820.
[3]	田文强, 董艳雪, 史永清, 雷钧杰, 孙刚刚, 王泓懿, 聂凌帆, 郭飞, 艾洪玉, 石书兵, 张金汕. 播期对超晚冬麦茎秆性状及群体动态的影响[J]. 新疆农业科学, 2023, 60(6): 1301-1307.
[4]	沈煜洋, 赖宁, 范贵强, 崔燕华, 杨红, 来汉林, 林敏, 雷钧杰, 李广阔, 高海峰. 小麦晚播和有机肥替代氮肥对麦长管蚜种群动态的影响[J]. 新疆农业科学, 2023, 60(5): 1082-1087.
[5]	王迎超, 张婧婧, 达新民, 耿新雪. 基于K-means与KNN的多特征苹果在线分级[J]. 新疆农业科学, 2023, 60(3): 643-650.
[6]	程文亮, 宋文钰, 王浩东, 许同心, 吴彩兰, 王春娟, 杨德松. 基于QuEChERS-GC/MS法对土壤和甜菜中精异丙甲草胺的残留降解动态测定[J]. 新疆农业科学, 2023, 60(3): 706-714.
[7]	陈玉兰, 程旭睿, 郭君, 苏武峥. 我国农业绿色发展水平地区差异及分布动态演进[J]. 新疆农业科学, 2023, 60(1): 252-260.
[8]	赵慧兵, 马娟, 张杰, 李浩, 郭磊, 冯斌. 肉牛自动化无应激动态称重系统的设计与试验[J]. 新疆农业科学, 2022, 59(7): 1795-1801.
[9]	张鲜花, 李江艳, 吴咏梅, 朱进忠. 镰芒针茅、博洛特绢蒿草地群落年际动态[J]. 新疆农业科学, 2022, 59(7): 1802-1807.
[10]	张鲜花, 李江艳, 吴咏梅, 朱进忠. 天山北坡中段山地紫花鸢尾草甸植物群落动态分析[J]. 新疆农业科学, 2022, 59(6): 1522-1529.
[11]	索银·图娅, 魏杨, 吾买尔江·巴哈提, 阿地力·沙塔尔, 张硕. 杏园梅下毛瘿螨Acalitus phloeocoptes(Nalepa)发生规律[J]. 新疆农业科学, 2022, 59(5): 1156-1164.
[12]	陆燕, 帕热古丽·卡看, 张蓓, 朱建梅, 阿地力·沙塔尔. 基于生物多样性指数分析沙棘林昆虫动态变化[J]. 新疆农业科学, 2022, 59(4): 966-976.
[13]	户金鸽, 白世践, 陈光, 赵荣华, 蔡军社. 不同级别无核白葡萄果实外观与品质差异特征及分级[J]. 新疆农业科学, 2022, 59(3): 567-577.
[14]	白剑宇, 罗达, 刘正兴, 李宏. 枣树花叶病毒病的分子检测[J]. 新疆农业科学, 2022, 59(3): 657-662.
[15]	阿米热·牙生江, 阿地力·沙塔尔. 白蜡窄吉丁成虫植物源引诱剂与粘虫板的筛选及发生动态分析[J]. 新疆农业科学, 2022, 59(10): 2532-2537.