基于免疫算法的自适应小波变换在红外图像去噪中的研究-生物医学工程专业论文.docx

基于免疫算法的自适应小波变换在红外图像去噪中的研究 基于免疫算法的自适应小波变换在红外图像去噪中的研究 RESEARCH ON INFRARED IMAGE DENOISING BASED ON IMMUNE ALGORITHM AND ADAPTIVE WA VELETTRANSFORMATION ABSTRACT Denoising infrared image is a kind of necessary pre-processing work for further image processing since many kinds of noise can be added into infrared image during the process of collection, transmission of Infrared image. There are a lot of kinds of noise in the infrared image. And the 2 main kinds of noise are the impulse noise and Gaussian noise. The median filtering is a nonlinear de-noising method and widely used to remove impulse noise. Wavelet transformation is a powerful tool for mathematical analysis. In recent years , wavelet transformation has been widely used in the field of infrared image denoising. There are many wavelet denoising algorithms which perform well to denoise infrared image with Gaussian noise. In this paper, we deeply investigate the method to denoise the impulse noise and Gaussian noise sine they are common in infrared image. The research work mainly includes 3 aspects as following: Through analyzing and studying denoising algorithms of time domain for infrared image, the several filtering algorithms are implemented to denoise infrared image. The experimental result shows that median filtering algorithm is performed better for impulse noise than other algorithms by comparison. After widely investigating image denoising method based on wavelet transformation and deeply analyzing the characteristics of image wavelet coefficients, a new kind of algorithm is proposed which combines the algorithms of time domain and frequency domain. This time- frequency domain filtering algorithm includes median filtering algorithm in time domain and wavelet filtering algorithm in frequency domain.The experimental result shows that the performance of this algorithm is better than that of wavelet filtering algorithm for Gaussian noise. The wavelet threshold plays a vital role in the wavelet threshold den