模糊神经网络像素分类的稀疏表示医学CT图像去噪方法

Research of Sparse Representation Medical CT Image Denoising using Pixels Classification by Fuzzy Neural Network

摘要: 在医学CT成像过程中，由于引入了不可避免的噪声，致使图像质量下降，影响临床诊断。因此，研究医学CT图像降噪方法在诊疗服务中具有重要意义。本文结合图像分割的思想，利用模糊神经网络将图像像素分成边缘区、平滑区与纹理区等不同图像区域，通过小波稀疏表示对不同类型的图像块进行阈值去噪处理，以便更好地保留医学CT图像的细节特征。实验结果表明，本文算法对医学CT图像降噪有一定的效果，峰值信噪比（PSNR）和结构相似性指数（SSIM）都得到了改善，更好并且很好地保留CT图像的细节信息。

Abstract: In medical CT imaging procedure, the unavoidable noise, results in image degradation, and has an influence on clinical diagnosis. Therefore, the study of medical CT image denoising method has great significance in the diagnosis and treatment services. In this paper, combined with the idea of image segmentation, image pixels are divided into edge region, texture smooth area using fuzzy neural network. Threshold denoising was present in wavelet sparse representation for different images. It better preserves the details of medical CT image. Experimental results show that The algorithm can effectively remove noise. The peak signal to noise ratio (PSNR) and structural similarity index (SSIM) have been improved. It well preserved edge details of the CT image.