Comparison of Image Enhancement and Sharpening Technique for Magnetic Resonance Image Quality Improvement

In medical imaging, clear and detailed images are important to help doctors detect and analyze diseases accurately. However, medical images such as magnetic resonance image (MRI) often have low contrast and unclear edges, which can make diagnosis difficult. To improve this, image enhancement and image sharpening techniques are used as preprocessing steps. Image enhancement and sharpening play a pivotal role in medical image processing, where the clarity and contrast of anatomical structures directly impact diagnostic accuracy. Image enhancement focuses on improving the overall brightness and contrast of an image to make hidden areas more visible. Meanwhile, image sharpening aims to highlight fine details and edges, making the boundaries between tissues clearer. This study compares both types of techniques—enhancement and sharpening—to understand their strengths and weaknesses when applied to MR images. Four image enhancement techniques are examined in this paper: Histogram Equalization (HE), Contrast-Limited Adaptive Histogram Equalization (CLAHE), Morphological Residual (MPR) and Type-II Fuzzy Set (T2FS) Contrast Enhancement. For image sharpening, five techniques are considered: Unsharp Masking, High-Boost Filtering, Laplacian Filtering, Nimble Filtering, and Adaptive Sharpening. Finally, the techniques are analyzed using both qualitative and quantitative methods. Based on the results, MPR and Nimble Filtering emerged as the top-performing techniques, yielding PSNR values of 45.0865 dB and 51.6507 dB, SSIM scores of 0.9712 and 0.9979, and MSE values of 0.0027 and 0.1515, respectively. These two approaches were subsequently integrated to form a hybrid method, which achieved superior results with a PSNR of 56.3271 dB, SSIM of 0.9994, and MSE of 0.0071. The proposed hybrid preprocessing strategy offers a promising preprocessing strategy with demonstrated improvement in MRI interpretation.