Variations of Adaptive Histogram Equalization (AHE) analysis on intra-oral dental radiograph

Abstract

Medical images, such as dental radiographs, suffer of low contrasts and noise which make it difficult to identify disease characteristics. This work utilizes the variation of Adaptive Histogram Equalization (AHE) towards enhancing the intra-oral dental radiograph. The algorithms used are AHE, Contrast Limited Adaptive Histogram Equalization (CLAHE), Sharpening Adaptive Histogram, Equalization (SAHE), Sharpening Median Adaptive Histogram Equalization (SMAHE) and Sharpening Contrast Limited Adaptive Histogram equalization (SCLAHE). These image enhancement algorithms variation are aim to test the noise, blurring and low contrast factors of the intra-oral dental image. The characteristic of the original and enhanced images are compared using image histogram. Results show that histogram of SCLAHE is the closest towards mimicking the original image histogram but with better pixel distribution, thus make the image visually clearer. The performance among the enhanced images is further compared using Contrast Improvement Index (CII), Signal to Noise Ratio (SNR) and Root Mean Square Error (RMSE). Results show that SCLAHE able to get the highest SNR, but the lowest CII and RMSE.