Study On Image Enhancement For BGA Ray Image

The BGA(Ball Grid Array)ray images have the characteristics of low signal-to-noise ratio,low contrast.BGA ray image enhancement processing is an important part of the BGA defect detection process,laying the foundation for the subsequent image segmentation and defect recognition.In this paper,concerning the characteristics of BGA ray image noise,the image enhancement algorithms based on BGA-ray image were researched,the main content includes the following aspects:1.A generalized unsharp mask algorithm for BGA ray image's contrast enhancement and sharpening was studied.Generalized unsharp mask algorithm using exploratory data model as a unified framework,can solve the problems of the halo effect in classic unsharp mask,as well as the gray-scale value out of range.The experiments show that the algorithm can significantly improve the the BGA-ray image's contrast and sharpness.2.P-M anisotropic diffusion model and its improvements was researched,improved anisotropic diffusion model not only consider the gradient of the image,and also take the gray variance into account.This method can remove noise,but also retained the edge of the image and details.The experimental results show that this improved anisotropic diffusion method can not only remove the noise effectively,but also retained noise BGA ray image's sharp edges and the tiny details.3.The image enhancement method based on mathematical morphology was discuss and an image enhancement algorithm of multi-scale image feature extraction based on top-hat transform was studied.This algorithm,using the multi-scale theory,can enhance image details under different scales.Using the top-hat transform,bright and dark areas and at each scale and details at adjacent scales were extracted.And then these features were used to enhance the contrast.The experimental results show that bright and dark areas and the details of the image are effectively extracted and processed.Thus,this algorithm can enhance the image,and suppress noise.