multiscaledbased

  第一：边界加强，通过估计噪声和实际信号的比值，来确定各个尺度的细节系数，以达到噪声抑制和边界加强。第二，通过对小波分解的近似部分进行直方图均衡，达到加大整幅图片的对比度，而不太影响细节的目的。(First: to strengthen the border, through the estimates and the actual signal noise ratio, to determine the details of each scale factor, so as to achieve noise suppression and boundary strengthening. Secondly, the wavelet decomposition of the approximate portion of histogram equalization to increase the contrast of the whole images, and do not affect the details of the purpose.)