基于小波变换统计模型的电子元器件图像去噪算法研究

在先进电子制造中,为了应用机器视觉方法来完成电子元器件的检测、处理和识别,必须对采集的相关图像进行去噪处理。文章研究了一种基于小波变换统计模型的去噪算法,利用四树复小波包变换把含噪图像高频方向子图分为主要类和次要类。然后,采用非高斯双变量模型和零均值高斯分布模型分别对主要类和次要类复系数进行建模,从而实现噪声抑制功能。实验结果验证了本文方法的有效性。

Statistical Model Based Wavelet Transform for the Image Denoising of Electronic Components

In advanced electronic manufacturing,the image de-noising processing is very necessary for the machine vision method,which can complete the testing,handling and identification of electronic components.In this paper,a novel image de-noising method based on wavelet packet transform is presented by using a mixed statistical model.The noisy image was decomposed into a low frequency approximation sub-image and some high frequency directional sub-images via the wavelet packet transform,and the high frequency directional sub-images are classified two categories：major coefficients and minor coefficients.So the noise in the major coefficients and the minor coefficients are removed by using of a mixed statistical model combining non-Gaussian bivariate model with zero mean Gaussian distributing model.The simulation has shown the effectiveness of the proposed method.