基于MATLAB的桑蚕茧选茧辅助检验方法的研究  

Study on auxiliary testing method for mulberry silkworm cocoon sorting based on MATLAB

作　　者：陈浩 杨峥 刘霞 邵玲 

机构地区：[1]山东省纤维检验局国家茧丝绸产品质量监督检验中心(山东),济南250032

出　　处：《丝绸》2016年第3期32-36,共5页Journal of Silk

摘　　要：针对目前桑蚕茧选茧均采用人工视觉进行判别易导致误选率高的问题,采用MATLAB软件对桑蚕茧彩色图像进行图像处理和数值计算的方法,从计算机视觉的角度,对桑蚕茧表面污斑面积的自动检测进行了分析,得到了一种智能化和自动化的选茧辅助检验方法。研究结果表明:在多光源拍摄彩色图像和进行准确相机标定的条件下,将桑蚕茧彩色图像经过彩色空间转换、图像增强和二值化等处理后,桑蚕茧表面污斑面积的自动检测快速且具有较高的准确度,颜色浅淡和多个污斑的样品同样具有良好的检测效果,所采集图像与实际大小之间的差异可采用立体相机进行三维模型重建来消除。Manual mulberry silkworm cocoon sorting may easily result in high false sorting rate. To solve this problem,MATLAB software was applied for image processing and numerical calculation of color images of mulberry silkworm cocoons. Automatic detection of the stained surface area of mulberry silkworm cocoon was analyzed from the perspective of computer vision,and an intelligent and automated auxiliary testing method for cocoon sorting. The results show that under the conditions of color image shooting with multiple light sources and accurate camera calibration,after color space transformation,image intensification and binaryzation of mulberry silkworm cocoon color image,the stained surface area of mulberry silkworm cocoon can be auto-detected accurately and rapidly. The samples with light color and multiple contamination spots also have the good detection effect. The difference between the gathered image and actual image can be eliminated through 3D modeling with a stereo camera.

关 键 词：桑蚕茧 选茧 MATLAB 计算机视觉 图像处理 相机标定 

分 类 号：TS101.8[轻工技术与工程—纺织工程]