图像处理在大米品质检测中的应用研究  

Study on Application of Image Processing on Rice Quality Detection

作　　者：马超[1] 周晚[2] 张志文[1] MA Chao;ZHOU Wan;ZHANG Zhiwen(School of Electronic Information Engineering,Xi’an Technological University,Xi’an 710021,China;Xianyang Vocational Technical College,Xianyang 712000,China)

机构地区：[1]西安工业大学电子信息工程学院,陕西西安710021 [2]咸阳职业技术学院,陕西咸阳712000

出　　处：《机械与电子》2018年第4期58-62,共5页Machinery & Electronics

基　　金：西安工业大学校长基金项目(XAGDXJJ15012)

摘　　要：基于图像处理技术建立了大米品质检测系统。利用直方图均衡化进行图像预处理,采用Canny算子检测边缘,通过与标准模板图像的比对,计算出待识别的大米图像与标准图像颜色、像素面积和纹理的差异性。结果表明,直方图均衡化有效增强了大米图像,图中的细节特征得以显现,Canny算子的边缘检测效果较好;黑米图像与标准图像的RGB距离最大;细长状的丝苗米和粘米与标准图像的像素面积差最大;各种米粒的纹理均有差异。实验结果与实际情况相符合,系统能够实现对大米外观品质的有效检测。A rice quality detection system was established based on the image processing technology.Firstly,the image pre processing was carried out through histogram equalization,and the Canny operator was used to detect the edge.Then,by comparing the standard template image,the differences in color,pixel area,and texture between the rice image and the standard image were calculated.The results indicate that histogram equalization can effectively enhance the rice image,the details of which can be revealed,and the Canny operator has good image edge detection effect.In addition,the RGB distance between the black rice image and the standard one is the largest.The slender silk rice and the sticky rice have the largest difference in pixel area from the standard image.The texture of the grains is different.

关 键 词：大米 图像处理 外观品质 颜色 面积 纹理 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]