多尺度塔型医学图像增强算法

在基于影像的医学诊断中，为了从图像中挖掘出尽可能多的决策信息，首先需要研究高效的图像增强算法。在分析人眼视觉特性和比较诸多图像增强方法的基础上，本研究选择并改进了一种多尺度塔型图像增强快速算法。以CR和DR临床图像增强为例，实验结果证明了该方法可以较好的满足人眼视觉主观要求，增强效果优于传统单一尺度方法。与小波多尺度图像增强相比，效果相当，但运算时间大为降低，具有良好的实际应用价值。

A Multiscale Pyramid Image Enhancement Algorithm and Its Application in Medical Image Processing

In the image-based medical diagnosis, it is required to have effective image enhancement algorithm for mining decision-making information as much as possible. On the basis of analyzing the human vision properties and comparing some available enhancement algorithms, a fast multiscale pyramid image enhancement algorithm was selected and modified in this paper. With computed radiography and direct radiography images as testbed, the experimental results illustrated that the proposed method could meet the subjective requirement of human vision well and exhibited better performance than traditional single-scale methods. Compared with the enhancement algorithms based on wavelet, the proposed method has advantage of less central processing unit time with similar enhancement effect.