摘要:
通过分析和研究二进提升方案,得到一种新的提升二进小波滤波器的构造方法.以新的B-样条二进小波作为初始二进小波滤波器,结合二维6 trous算法对原始医学图像进行多层分解,根据分解后系数特点,一方面对不同级数的高频系数利用单阈值和双阈值函数增强,对低频子带系数用线性拉升来扩大图像的整体对比度.另一方面结合消失矩条件,调整滤波器的提升参数,构造了具有线性相位、紧支撑、高阶消失矩及有限长的单位脉冲响应的提升二进小波,并将其与初始二进小波滤波器增强后图像进行比较.仿真结果表明提升后的滤波器对于对比度低、弱光照图像比初始滤波器的增强效果有了进一步提高,噪声放大问题有明显改善,更好地突出图像边缘特征、保留了图像更多细节信息.%Through analysis and research into the dyadic lifting scheme,a new construction method for lifting dyadic wavelet filter is proposed.With a new B-spline wavelet as the initial dyadic wavelet filter based on 2-D àtrous algorithm,the original medical image is decomposed,according to the decomposition coefficient characteristics,on the one hand,the high frequency coefficient different series using single threshold and double threshold function enhancement,the low frequency subband coefficients using linear transformation to expand the overall contrast of the image.on the one other hand,combined with vanishing moment conditions,enhance the adjustment of the filter parameters,constructed with linear phase,short support,high vanishing moments and limited long unit impulse response of the lifting dyadic wavelet,and the image is compared with the initial enhancement dyadic wavelet filter.The simulation results show that the improved filter for low contrast,weak illumination image results than the initial filter enhancement effect,the problem of noise amplification is obviously improved,the edge feature of image is better highlighted,and more details of image are reserved.