结合形状先验的水平集印刷电路板CT图像分割方法

PCB CT Image Segmentation Based on Level Set with Shape Prior

用CT对印刷电路板(PCB)进行无损检测是近年来发展起来的一种新方法.由于PCB中存在大量的金属元器件,在成像时产生较为严重的金属伪影,导致图像灰度不均匀问题较为严重.为此,提出一种结合形状先验的水平集PCB CT图像分割方法.首先根据PCB CT图像中导线具有明显的方向特征对图像进行不同方向的Gabor滤波并对结果加权求和,获得边缘增强的滤波结果,并通过局部化方法对边缘增强的结果消除背景噪声,经过阈值化处理和形态学闭运算后获得图像形状先验;然后用形状概率表示方法来表示形状先验,并将其作为形状约束项,与CV模型能量项、局部能量一起构成能量函数;最后通过对能量函数最小化实现图像分割.实验结果表明,该方法对多目标及严重灰度不均匀的PCB CT图像具有较好的分割结果.

It is a new method to nondestructively test PCB using CT in recent years. However, because of a lot of mental components in PCB, there will be serious metal artifacts, which results in the images with in-tensity inhomogeneity. Aiming at this problem, a novel level set method based on shape prior is proposed. Firstly, PCB CT image is filtered by different orientation Gabor filters according to the obvious orientation feature of lines in PCB. The edge-enhanced result can be obtained by weighted sum of different orientation filtered results and the background noise of edge-enhanced image can be weakened by localization method. The shape prior can be obtained after using threshold and morphological closing operation. Then, the energy function consists of Chan-Vese item, localized energy item and shape prior item which is represented by probabilistic definition of shape. Finally, the segmentation result can be obtained by minimizing the energy function. The experiments of PCB CT images with multi-object and intensity inhomogeneity demonstrate the performance of our model.