首页 | 本学科首页   官方微博 | 高级检索  
     

基于图像识别的整精米自动检测研究
引用本文:于润伟,金鲲鹏,朱晓慧. 基于图像识别的整精米自动检测研究[J]. 中国粮油学报, 2006, 21(6): 147-150
作者姓名:于润伟  金鲲鹏  朱晓慧
作者单位:黑龙江农业工程职业学院,哈尔滨,150088
摘    要:本文采取动态阈值分割,把米粒图像变成二值图像,根据区域内部像素的连通性,将不同的区域赋予不同的标记,计算出单个米粒所占的像素,再通过先期计算的整精米长度/面积比换算出米粒长度,最后根据米粒长度判断整精米和碎米。研究结果表明,该算法的自动检测与人工检测相关性大于99%,可用于整精米的自动检测。

关 键 词:整精米  图像识别  自动检测
收稿时间:2005-11-18
修稿时间:2005-11-18

Head Rice Automatic Detection System Based on Image Recognition
Yu Runwei,Jin Kunpeng,Zhu Xiaohui. Head Rice Automatic Detection System Based on Image Recognition[J]. Journal of the Chinese Cereals and Oils Association, 2006, 21(6): 147-150
Authors:Yu Runwei  Jin Kunpeng  Zhu Xiaohui
Affiliation:HLJ Agricultural Engineering Trade College, Harbin 150088
Abstract:The following is the presented detection procedure in this article: Adopt dynamic threshold disintegrating and change the head rice into binary images;according to the connection of the internal pixel,attach different labels to different regions and calculate the pixel which one head rice takes;then through the previous length/area ratio of head rice get the length;finally recognize the head rice and broken rice.The datum shows that the correlation between the results from this automatic detection process and manual detecting system is over 99%.This detection procedure surely can be used in the automatic detecting system.
Keywords:head rice   image recognition   automatic detection
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号