摘要
提出了一种基于数学形态学的失焦检测方法。对一组工业模具用针分别进行准确对焦以及失焦拍照,通过Kmeans聚类原理将图像转化为二值图像,采用数学形态学的方法对目标图像进行先膨胀后腐蚀的处理,改善图片质量,获取更为准确的边界信息。以目标图像特征提取的方法筛选出面积低于阈值面积的模具用针图像,从而确定已发生断裂的针。通过对比不同焦距照片的处理结果,提出失焦检测方法,明显提高检测效率。实验证明,此方法可以快速、准确、批量的检测出模具断针,确立适宜焦距,在工程上具有广泛的应用前景以及实用价值。
A kind of out-of-focus detection method based on mathematical morphology is proposed. Take accurate focus and out-of-focus pictures of a group of pins on mould and convert the picture to binary image by the k means clustering method. To improve the quality of the image and acquire the more precise edge, dilate and erode the image through the mathematical morphology method. By the method of screening the pin which image area is below the threshold, the damaged one can be picked. By comparing the results of different focal length,a kind of out-of-focus detection method is proposed and the efficiency can be raised. The experiment proves that this method can quickly and precisely detect the damaged pin in batch and determine the optimum focal length. This method is of great practical value and has promising application prospect.
出处
《机械》
2016年第6期67-70,77,共5页
Machinery
基金
航天医学基础与应用国家重点实验室研究基金资助(SMFA12B03)
载人航天领域预先项目资助(040101)
关键词
工业损伤
数学形态学
Kmeans聚类法
二值化
失焦检测
industrial damage
mathematical morphology
Kmeans clustering method
binarization
out-of-focusdetection
