摘要
建立一种基于色相-饱和度-明度(HSV)颜色空间的二值化方法及多种颜色空间模型的稻谷脱壳率的检测方法。方法先采用颜色空间变换和阈值法对图像的二值化进行处理,然后采用轮廓检测算法得到稻谷的外接矩形框,再采用对比分析法从图像的红、绿、蓝色的灰度(RGB)和HSV模型中提取R值和H值作为分类特征;最后采用支持向量机和K均值聚类算法处理特征数据,得到稻谷脱壳率。检测方法能够有效地完成稻谷脱壳率的检测任务。
A method was established for the detection of rice husking rate based on the binarization method of hue-saturation-value(HSV)color space and multiple color space models.The method first used the color space transformation and threshold method to process the binarization of the image,then used contour detection algorithm to obtain the outer rectangular box of the rice,and then used compara-tive analysis method to extract R and H values as classification features from the red,green,and blue gray scale(RGB)and HSV models of the image.Finally,support vector machine and K-means cluste-ring algorithm were used to process the feature data and obtained the rice husking rate.The detection method can effectively complete the task of detecting the husking rate of rice.
作者
任建新
张士雄
李昂
任瑞龙
REN Jian-xin;ZHANG Shi-xiong;LI Ang;REN Rui-long(School of Mechanical and Electrical Engineering,Henan University of Technology,Zhengzhou 450000,Henan,China)
出处
《粮食与油脂》
北大核心
2023年第12期154-157,162,共5页
Cereals & Oils
基金
河南省科技攻关计划项目(222103810085)。
关键词
二值化方法
颜色空间
支持向量机
聚类算法
脱壳率
binarization method
color space
support vector machine
clustering algorithm
husking rate
