Medical Image Fusion Employing Enhancement Techniques

摘要

The contrast and quality of medical images get degraded due to the inherent properties of an imaging system which leads to inaccurate diagnosis. Nevertheless, such problems can be rectified by image enhancement and fusion methods. On this account, the paper encompasses the efficacy of four widely used enhancement techniques, namely, Binarization, Median filter, Contrast Stretching (CS), and Contrast Limited Adaptive Histogram Equalization (CLAHE) using two conventional (Principal Component Analysis, Discrete Wavelet Transform) and a hybrid fusion technique. To evaluate the performance of the considered algorithms, Peak Signal to Noise Ratio (PSNR), Signal to Noise Ratio (SNR), and Structural Similarity Index Measure (SSIM) are some of the performance metrics considered. From the experimental findings, it is observed that CLAHE outperforms other methodologies. For the proposed hybrid method using CLAHE, 0.72, and 0.50 SSIM values are obtained for reference and fused images respectively which results better in contrast to CS (0.58 and 0.26 respectively SSIM).