基于苹果着色面积的计算机视觉分级技术研究  

Study on computer vision grading based on apple coloration area

作　　者：王建[1] 黎绍发[2] 

机构地区：[1]华南农业大学工程学院,广东广州510642 [2]华南理工大学计算机科学与技术学院,广东广州510641

出　　处：《计算机工程与设计》2008年第14期3813-3814,3817,共3页Computer Engineering and Design

基　　金：广东省工业攻关计划基金项目(2004B10101032);华南农业大学博士启动基金项目(B2004003)

摘　　要：通过数码相机获取不同苹果等级的彩色图像,将图像平滑后利用R颜色分量进行最小错误阀值分割,再进行腐蚀去噪声和孔洞填充,并将RGB值转换成HSL值,最后用像素点变换法恢复苹果在二维投影图中真实的几何信息,重新计算像素点的面积,计算苹果的着色面积并进行分级。实验结果表明,机器识别苹果等级的准确率为89.8%。The different quality level apple color pictures are taken by digital cameral and processed in the grading system based on Visual C＋＋ 6.0 developed. Firstly, using R color branch to determine threshold value after smoothing image, then the image segment and erosion are carried through, Finally, the image RGB is changed to HSL, and the true apple area is obtained by means ofpixel points transform method taking count of the number of pixels on the projection. The results show that the method is proper for the apple sorting based on computer vision, and it has a result with a veracity of 89.8% comparing with manual judgment.

关 键 词：着色面积 像素点变换法 机器视觉 分级 计算机应用 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]