Fitness Function Evaluation for Image Reconstruction using Binary Genetic Algorithm for Parallel Ray Transmission Tomography

Various fitness functions have been evaluated for image reconstruction using binary genetic algorithm (BGA) based parallel ray transmission tomography. The population initialization is carried out using the filtered backprojection (FBP) technique. Various fitness functions used for image reconstruction include: root mean squared error (RMSE), mean squared error (MSE), mean absolute error (MAE), relative squared error (RSE), root relative squared error (RRSE) and relative absolute error (RAE). RMSE and MAE outperformed for small as well as large size images with different shape complexities. Mixed selection scheme with two variations of crossover operators, namely image-row and block crossover operators have been used for crossover. Binary mutation operator has been used for creating diversity in local search scope. For 64 times 64 head and lung phantoms, BGA has resulted in PSNR values with RMSE 19.26 and 16.49 respectively and 27.20 and 29.65 dB with MAE