基于多尺度分析的血管内超声图像去噪和增强

提出基于二进小波变换的血管内超声图像血液斑点噪声抑制和对比度增强算法.血液红细胞散射引起的斑点噪声属于乘性噪声,在对数域进行二进小波变换后,结合软阈值滤波法和硬阈值滤波法对不同尺度的小波系数进行萎缩处理,并提出了一种局部阈值估计方法.同时采用了基于多尺度边缘表示的,利用小波系数极值拉伸和Hermite多项式插值实现的快速增强算法.实验结果表明,与现有单独进行去噪处理的方法相比,该方法在抑制血液斑点噪声的同时增强了图像对比度,具有更好的实用性.

Denoising and enhancement in intravascular ultrasound images via multiscale analysis

An algorithm based on dyadic wavelet transform for speckle reduction and contrast enhancement in intravascular ultrasound images was presented. Wavelet shrinkage techniques which combined soft and hard thresholding were applied to coefficients of logarithmically transformed images since scattering from red blood cells (blood speckle noise) was multiplicative noise and a method to estimate local threshold was proposed. In addition, a fast contrast enhancement algorithm based on multi-scale edges representation of images through stretching the local extrema and interpolating them with Hermite interpolation polynomials was carried out. Experiments with clinical images showed that this algorithm was capable of not only reducing the speckle noise of blood but also enhancing features of diagnostic importance of intravascular ultrasound images and produced superior results qualitatively when compared to results obtained from existing denoising methods alone.