摘要
研究基于小波变换的隐式马尔科夫模型树(HMT)方法,用于去除图像信号中的白噪声。该方法利用了小波变换域系数间的相关性和自相似信息,并在Lenna图像中验证了方法的有效性。对不同程度污染的高斯白噪声图像的去噪效果与传统方法进行比较;结果表明,基于小波变换的HMT方法有更好的去噪效果,所建立的HMT模型更能体现图像的特征。
Wavelet-domain Hidden Markov Tree（HMT）method have recently been studied and applied to image processing, The advantage of the method is that the HMT measure the dependency and self- similarity inherent in transformed images. The effectiveness of the method is demonstrated by using numerical simulations in the image of Lenna, Comparing with traditional denoising method into different White Gaussian noise images,the results show that the HMT method is better in denoising and enhancing signal -to -noise ratio and the HMT model embodies the characteristic of image.
出处
《现代电子技术》
2006年第2期21-23,共3页
Modern Electronics Technique
基金
国家863高科技计划项目(2002AA412120)
河南省青年骨干教师项目资助
