基于广义高斯分布模型的基因芯片图像去噪

基因芯片图像去噪是基因芯片应用过程中一个非常重要的步骤,对于芯片数据处理和信息提取具有重要意义.用广义高斯分布对芯片图像子带的小波系数进行建模,在此基础上运用Bayes Shrink法对图像进行小波去噪.实验结果表明,这种方法能够在有效去除基因芯片图像噪声的同时,很好的保持图像的边缘,与其它几种去噪方法相比,不仅提高了去噪后图像的信噪比(SNR)和均方误差(MSE),而且使图像更加清晰,为芯片数据进一步的分析处理奠定了基础.

Microarray Image Denoising Based on Generalized Gaussian Distribution Model

Microarray image denoising is a significant step in the process of microarray application, and is of great importance for microarray data processing and information extraction. In this paper Bayes Shrink is used to remove noise based on the generalized Gaussian distribution statistical model for wavelet coefficients of microarray image. Experiment results show that this method can effectively suppress noise and preserve edge signal information of microarray image comparing with other conventional methods, which not only improves the SNR(Signal-to-Noise Rate) and MSE(Mean Squared Error), but also makes denoised image more clear. The method lays the foundation for further microarray data analysis.