An Improved Fusion Technique for Medical Images Using Discrete Wavelet Transform

Nowadays, most of the physicians face the problem of diagnosing and directing the patients with a single image solution. This scenario arises when they have to diagnose the disease with the Positron Emission Tomography (PET) image, which has poor spatial resolution. This constraint is overcome by integrating the PET image with Magnetic Resonance Image (MRI), which is rich in spatial resolution. This paper presents a novel approach to image fusion based on two processes, namely, (i) image enhancement and (ii) image fusion. These two processes will plunge on traditional Discrete Wavelet Transform (DWT). In this work, the PET image is enhanced using the DWT and then combine with the MRI image. The evaluation of the proposed system is conducted on two different scenarios namely simulation and implementation. In the simulation the visual comparison and quantitative analysis such as discrepancy and average gradient are computed to validate the spectral and spatial resolution of the fused image. Also, the Peak-Signal-to-Noise-Ratio (PSNR), Mean Square Error (MSE) and Entropy of the proposed image fusion method are compared with various traditional methods such as, Principle Component Analysis (PCA), Intensity-Hue-Saturation (IHS), Wavelet and Curvelet. In contrast, the FPGA implementation of the proposed method is evaluated in terms of Logic Elements (LE's), power and maximum frequency. In the entire evaluation scenario, the proposed image fusion method shows a remarkable improvement.