农作物状态的三色信息边缘检测研究  

On Tricolor Information Edge Detection of Crop State

作　　者：邓蕾蕾[1] 张献 王政皓 黄涛 王闯 DENG Lei-lei;ZHANG Xian;WANG Zheng-hao;HUANG Tao;WANG Chuang(College of Information Technology, Jilin Agricultural University, Changchun 130118, China;School Office, Changchun University, Changchun 130022, China)

机构地区：[1]吉林农业大学信息技术学院,长春130118 [2]长春大学学校办公室,长春130022

出　　处：《西南师范大学学报（自然科学版）》2020年第9期59-65,共7页Journal of Southwest China Normal University(Natural Science Edition)

基　　金：吉林省教育厅课题(JJKH20180650KJ).

摘　　要：对于农作物状态检测的传统方法是人工检测、光谱检测等,普遍存在检测准确率不高和检测速度较慢的问题.本文以摄像机为传感器对农作物进行现场实时拍摄,进而采用一种三色信息边缘检测的方法做出实时判断.在这个检测方法中,农作物图像首先被分解为R(Red),G(Green),B(Blue)这3个分量信息图像.然后,在R,G,B这3个分量图像上,农作物的颜色信息和边缘信息被提取出来,边缘信息提取采用的是Roberts算子.最终,通过3个颜色信息特征和3个边缘信息特征构建出相似性测度,作为检测结果的判断依据.针对西瓜样品图像展开实验研究,实验结果显示:本文方法可以获得丰富的特征信息,包括西瓜的3个颜色特征信息和3个边缘特征信息,据此得出的品质判断结果准确.对比实验表明:当样本数量为500个时,本文方法的检测准确率仍然达到了96%.The traditional method of crop condition detection is manual detection,which generally has the problems of low detection accuracy and slow detection speed.Therefore,in this paper,the camera has been taken as the sensor to take real-time photographs of crops,and then a method of tri-color information edge detection used to make real-time judgment.In this detection method,crop images have first been decomposed into R,G and B component information images.Next,the crop color information and edge information have been extracted from the R,G and B component images.Roberts operator has been used to extract the edge information.Finally,the similarity measure has been constructed by three color information features and three edge information features,which can be used as the judgment basis of the test results.Experiments on watermelon sample images show that this method can obtain rich feature information,including three color feature information and three edge feature information of watermelon,and the quality judgment results are accurate.

关 键 词：农作物状态 三色分量信息 边缘特征 ROBERTS算子 

分 类 号：S5-39[农业科学—作物学]