基于分块阈值LBP算法的光学薄膜表面缺陷分割

Segmentation of optical thin film surface defects based on LBP algorithm with block threshold values

为提高光学薄膜缺陷检测的能力和精度,提出一种基于局部二值模式(LBP)的改进算法。通过对采集到的图像进行分块分析,计算全图和各小块离散度,并进行对比。将分块离散度>全图离散度的视为缺陷与背景共存的模块,以该模块中间像素灰度均值作为LBP阈值进行图像处理;分块离散度≤全图离散度的视为平稳模块,将全图灰度均值作为LBP阈值进行图像处理。实验结果表明:该算法能有效地检测出光学薄膜的缺陷,同时也避免了噪声的干扰,证明该算法用于检测光学薄膜缺陷的可行性。

To improve the ability and accuracy of defect detection of optical thin films,an improved algorithm based on local binary patterns(LBP)was proposed.The image was divided into blocks and computed the dispersion of each small block and the whole image for comparison.The module with block dispersion higher than global dispersion was regarded as the module with defect and background coexistence,using the average gray value of the middle pixel as the threshold value for LBP processing.The module with block dispersion lower than and equal to global dispersion was regarded as the stable module,using the average gray value of the whole image to process these module.The experimental results showed that this algorithm could effectively detect the defects of optical thin films and avoid noise interference at the same time,and the algorithm was feasible for optical thin film defect detection.