基于颜色和纹理特征的异常玉米种穗分类识别方法  

Classification and recognition method of abnormal corn ears based on color and texture features

作　　者：李颀[1] 王康 强华 马琳[1] LI Qi;WANG Kang;QIANG Hua;MA Lin(College of Electrical and Control Engineering,Shaanxi University of Science and Technology,Xi′an 710021,China)

机构地区：[1]陕西科技大学电气与控制工程学院

出　　处：《江苏农业学报》2020年第1期24-31,共8页Jiangsu Journal of Agricultural Sciences

基　　金：陕西省科技厅农业科技攻关计划项目(2015NY028)

摘　　要：针对玉米选种过程中异常种穗的外观缺陷难以准确识别的问题,以玉米种穗为研究对象,通过计算机视觉技术快速识别杂色、缺粒、虫蛀、籽粒杂乱4种异常种穗。选择单目视觉采集装置,采集任意姿态玉米种穗图像,利用凹点匹配算法分割粘连玉米种穗;采用HSV和CLBP(完全局部二值模式)方法提取玉米种穗的颜色和纹理特征,利用匹配得分融合算法融合玉米种穗的颜色和纹理特征,建立玉米种穗分类模型,利用SVM实现4种异常玉米种穗的快速分类。试验结果表明,该方法相对于传统玉米种穗检测技术能快速有效识别出4种异常玉米种穗,对杂色、缺粒、虫蛀、籽粒杂乱玉米种穗的识别正确率分别达到了96.0%、94.7%、93.6%和95.3%,玉米种穗在有粘连和无粘连情况下平均识别速度分别为每穗1.180 s和0.985 s,能够满足异常种穗分类识别的需求。This study aims at the problem that it is difficult to accurately identify the appearance defects of abnormal corn ears during corn selection.Taking the whole corn ear as the research object,the four kinds of abnormal corn ears(variegated corn ear,missing corn ear,worm-eaten corn ears and untidy corn ears)were quickly identified by computer vision technology.The monocular visual image acquisition device was selected to collect the image of the corn ear in arbitrary posture,and the pit matching algorithm was used to complete the rapid segmentation of the cohesive corn ear.The HSV color model and complete local binary pattern(CLBP)method were used to extract color and texture features of corn ear,and the matching score fusion algorithm was used to fuse the color and texture features of corn ear.At the same time,an abnormal corn ears classification model was established.Finally,the rapid classification of four abnormal corn ears was achieved by support vector machine(SVM).The experimental results showed that this method could quickly and effectively identify four abnormal corn ears compared with traditional corn ear measurement technology.The correct recognition rate of this method was 96.0%,94.7%,93.6%and 95.3%for the variegated corn ears,missing corn ears,worm-eaten corn ears and untidy corn ears.The average recognition speed of maize seed ear in the presence of adhesion and non-adhesion was 1.180 seconds per ear and 0.985 seconds per ear.This method can provide a basis for the identification of abnormal ears during the intelligent corn ear sorting process.

关 键 词：玉米种穗 计算机视觉 颜色特征 纹理特征 分类识别 

分 类 号：S126[农业科学—农业基础科学] TP391.41[自动化与计算机技术—计算机应用技术]