Efficient image fusion using multi scale decomposition and absolute maximum fusion rule for MRI and CT brain images

Currently, multimodal image fusion has evolved to be an exciting topic in medical technological field. If the multimodal images are directly applied into use, there will be a huge increase workload pertaining to diagnosis of disease, incorrect decisions to occur. However, the existing method has issue with inaccuracy image results and computational overhead. However, blur detection still suffers from problems such as the oversensitivity to image noise and the difficulty in cost-benefit balance. To solve these issues proposed work introduced an EBO + AMFR (Enhanced Bat Optimization and Absolute Maximum Fusion Rule). Proposed work major steps are pre-processing, decomposition, feature extraction, feature selection and image fusion process. Initially pre-processing, improving image quality through elimination of artifacts. Secondly, CTS and MRI images are split into lower, higher layers based on their frequencies applying linear filter based multi scale decomposition process. Thirdly, feature extraction is applied through MPCA (Modified Principal Component Analysis) which extracts more informative image features. Then, feature selection process applying EBO (Enhanced Bat Optimization) to selects more relevant, important features from given image database. Finally, image fusions using AMFR (Absolute Maximum Fusion Rule) which merges useful, significant image features. The performance evaluation metrics are accuracy, PSNRs (Peak Signal to Noise Ratios), RMSEs (Root Mean Square Errors) and execution times, the best balance between cost and benefit.