Histogram modification framework and its application to brain tumor segmentation

Abstract

Medical images help doctors diagnose the patients health conditions and treat them accordingly. Medical images with contrast issues due to insufficient lighting are inadequate for image analysis. In order to resolve the contrast issues, pre-processing of such medical images, which is significant, needs to be done before carrying out segmentation, feature extraction and classification operations. The primary task of pre-processing is to improve the quality of the MR images and to make it in a form suitable for further processing by human or machine vision system. Contrast enhancement is a significant pre-processing for medical image analysis. The histogram-based contrast enhancement methods treat the images as regions rather than objects, which would be more useful for applications like brain image enhancement. The image enhancement techniques are widely used in the field of medical imaging especially in Magnetic Resonance Imaging (MRI). MR images are used to scan the brain s internal regions. Basically, the MRI technique of imaging helps in generates good quality images of the human body parts. MRI is used for many purposes such as diagnosing brain tumors, multiple diseases, spinal infections; to visualize shoulder injuries, tumors in bones, and strokes. This research work mainly focuses on the histogram modification framework for image contrast enhancement. Histogram Equalization (HE) is one of the most popular techniques used for image contrast enhancement as HE is computationally fast and simple to implement. HE performs its operations by remapping the gray levels of the image on the basis of the probability distribution of the input gray levels. However, HE is rarely employed in medical image processing as it tends to introduce some annoying artifacts and unnatural enhancement with intensity saturation effect. One of the reasons for this problem is that HE normally changes the brightness of the image significantly and thus making the output image saturated with very bright or dark intensity values. A Fuzzy a