An Efficient Image Reconstruction Technique with Aid of PSO (Particle Swarm Optimization) and DWT (Discrete Wavelet Transform)
Bangaru Raju, Priyanka Pandarinath, G. S. Prasad · International Review on Computers and Software (IRECOS) · 2013
Image Reconstruction is to retrieve the original image (or a general signal) from its given awful version, for e.g., an image that is corrupted by noise, blurred by atmospheric turbulence (as in certain astronomic observations), or that has some scratched regions. Different reconstruction methods were utilized for performing the image reconstruction process. In such works, there is a lack of analysis in considering the reconstructed image quality because the reconstructed image seems to be blurred and poor in quality and so yielded less accuracy in the image reconstruction process. So avoid such drawbacks in the existing methods a new image reconstruction technique is proposed in this paper. The proposed technique comprised of two major phases (i) training phase (ii) investigation phase. In training phase, the given cracked image is reconstructed by the DWT (Discrete wavelet Transform) method by selecting optimal threshold value using PSO (Particle Swarm Optimization). These selected threshold values are exploited in the image reconstruction process. In investigation phase, the threshold value is selected based on the crack level of the testing image. By combining the DWT and PSO optimization in the proposed technique, the reconstructed image is obtained with high quality. The implementation result shows the effectiveness of proposed image reconstruction technique in reconstruct the image with different crack variance. The performance of the image reconstruction technique is evaluated by comparing the result of proposed technique with the average filtering image reconstruction technique. The comparison result shows a high-quality reconstructed image for the noisy images than the existing method, in terms of peak signal-to-noise ratio (PSNR).