• Mr. Ankit Kumar Srivastava
• Miss Pratiksha Rai

Abstract This paper aims in presenting a thorough assessment and utility of multi-resolution based image restoration techniques for object identification. Multi-resolution based image restoration techniques overcome the drawback of Fourier, spatial, as well as, purely frequency based techniques, as it provides the information of image at different resolutions and scales effectively. The multi-resolution based techniques have been implemented for the restoration of camera images. Further, the effectiveness and suitability of the techniques have been compared in terms of different datasets, noise and blur variance, subjective and by using well defined evaluation metrics such as, PSNR, RMSE and CC. Analysis of result shows that shift-invariant and multi-directionality NSCT technique not only extract more visual information from contaminated images, however, effectively avoid the introduction of artificial information. It significantly outperforms the SWT and DWT techniques in terms of object identification..

Image restoration, multi-resolution analysis, shift-invariant, Non Sub-sampled Contourlet Transform

"Image Restoration Using Multi - Resolution Based Techniques For Object Identification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.292-302, May-2019, Available :http://www.jetir.org/papers/JETIR1905943.pdf

"Image Restoration Using Multi - Resolution Based Techniques For Object Identification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp292-302, May-2019, Available at : http://www.jetir.org/papers/JETIR1905943.pdf