Multi-Range Fusion for X-ray Image Enhancement

Abstract

Digital medical images are stored in the Digital Imaging Communications in Medicine (DICOM) standard where a 12-bit integer represents each pixel. However, commercial display devices are only designed to work with an 8-bit standard such as JPEG or PNG. Traditional methods compress the high dynamic range (HDR) to a low dynamic range (LDR) to visualize a DICOM image. The compression process is a linear transformation and the LDR is represented in an 8-bits format. Due to a quantization error the contrast is limited. Therefore, in this article, we propose a method to enhance X-ray images in term of contrast. Firstly, we estimate an information range that captures valuable information of an x-ray image. The range is represented by an upper and lower threshold. Within these thresholds, LDR images will be extracted to capture information from a specific sub-range. These images are then fused to create an enhanced LDR image that has a higher contrast but retains information from each sub-range. Experimental results demonstrate that the proposed method helps achieve a better contrast.