Performance evaluation of a hybrid histogram equalization-contrast limited adaptive histogram equalization technique for medical images enhancement

The enhancement of medical imagery is crucial for accurate diagnosis by improving visual quality. Conventional techniques, however, often introduce undesirable effects like noise amplification, artefacts, and inadequate feature representation. To overcome these limitations, researchers have developed hybrid frameworks that integrate multiple methods. Examples include a trimodal technique that combines Unsharp Masking (UM), Adaptive Histogram Equalization (AHE), and Logarithmic Transformation (LT) and the fusion of Contrast-Limited AHE (CLAHE) with a Fuzzy inference system for contrast enhancement. Despite their advantages, these hybrid models suffer from excessive contrast enhancement, and sub optimal performance. This paper presents a performance evaluation of a novel hybrid framework that couples Histogram Equalization (HE) with CLAHE techniques, designed to enhance the visual quality and diagnostic utility of medical images. The proposed method, referred to in this study as hybrid HE-CLAHE technique, is developed by integrating the strengths of HE and CLAHE. The performance of the hybrid model is assessed using standard quantitative metrics, namely the Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM). İt is benchmarked against standalone HE, CLAHE, and relevant existing hybrid models from the literature. The results demonstrated that the hybrid HE-CLAHE method achieved the lowest MSE and highest PSNR and SSIM values across all tested medical image datasets (Computed Tomography (CT), X-ray, and Magnetic Resonance Imaging (MRI)). For illustration, when enhancing a human knee X-ray image, the hybrid technique yielded a PSNR of 26.80 dB, compared to 25.83 dB for HE and 25.94 dB for CLAHE. This corresponds to a performance gain of 3.74% over HE and 3.29% over CLAHE. Furthermore, the hybrid technique's PSNR performance was competitive with prior, more complex trimodal approach that integrated UM, AHE and LT techniques, validating the efficacy of the proposed simpler framework