An Analysis of Image Enhancement Techniques for Dental X-ray Image Interpretation

Abstract

This paper presents qualitative and quantitative comparison between original images and four image enhancement techniques namely adaptive histogram equalization (AHE), contrast adaptive histogram equalization (CLAHE), median adaptive histogram equalization (MAHE) and sharp contrast adaptive histogram equalization (SCLAHE) applied to dental x-ray images. Dental x-ray images usually taken with low radiation dosage are often presented as dark, low in contrast and noisy. These problems are usually solved with image enhancement techniques. However, choosing an appropriate technique is not an easy task especially for the purpose of disease diagnosis of periapical related lesion. This research involves the collection of ten intra-oral dental x-ray images which were collected from the Faculty of Dentistry UiTM Shah Alam, Malaysia. Each of the enhancement methods is applied to every collected image. A dentist was then asked to do the evaluation using questionnaire. She graded the quality of the images and the diagnostic ability of the periapical pathology observed in the images. Subsequently the quantitative analysis using contrast improvement index (CII), signal to noise ratio (SNR) and root mean square error (RMSE) were done to investigate the characteristic of the images base on the dentist evaluation. The finding shows that the enhancement techniques managed to enhance the pathology slightly better than the original image. Index Terms—Contrast enhancem