An Improved Medical Image Denoising Algorithm Based on One-Dimensional Heat Transfer Equation
YanZhu Zhang, MingHai Zhang, Qi Yang, Tianhao Wang · 2018 IEEE 7th Data Driven Control and Learning Systems Conference (DDCLS) · 2018
Denoising is a critical step for medical image processing. When applied to medical image processing, the traditional denoising algorithm has the disadvantage of being vague. This paper presents an improved image denoising method to combine the fractional differential mask operator and one-dimensional heat transfer equation. Due to the amplitude-frequency characteristic of fractional differential operation, this algorithm can preserve more image texture information and overcome the staircase effect in the region where the gray level of image smoothing does not change much. The algorithm has strong ability to remove noise and preserve the edge features and texture details of the image. The experimental results show that the medical images processed by the algorithm preserve more pathological information than that of the common method of denoising partial differential images. The improved algorithm provides reliable evidence for the subsequent medical diagnosis.