Moment-based image enhancement for brain tumor health monitoring

Since the stable increasing incidence of brain tumors in recent years, brain tumor detection and monitoring are being attached with more impor tance. To implement the image feature extraction approach for the current imaging system, the image mo- ments' concepts are introduced. The theory of image moments is applied for brain image analysis, which is a weighted average of the image pixels' intensities representing the characteristics of the mentioned brain images with potential tumor diseases. This paper describes several continuous and discrete moments in terms of the polynomial kernels used and distinguishes their differences regarding image recon struction and enhancement. The experimental results confirm that the proposed discrete Tchebichef and Krawtchouk moments are more robust in terms of noise and blur reduction than the existing methods, such as the Wiener filter. This process explains how th e proposed image moments technique can be applied in the health monitoring of brain tumors via image analysis procedures.