Anisotropic Image Restoration Based on Image Inpainting with Diffusion Enhancement
Marlapalli Krishna1, V Naga Bushanam2, Bandlamudi S B P Rani3, K Rakesh4, V Pranav5 

1Dr. M. Krishna, Professor, Department of Computer Science Engineering, Sir C R Reddy College of Engineering, India.
2V Naga Bushanam, Associate Professor, Department of Computer Science Engineering, Chirala Engineering college, India.
3Bandlamudi S B P Rani, Obtained M. Tech from Andhra Univrsity., India.
4K Rakesh, Assistant Professor, Department of Computer Science Engineering, GVPCDPGC, India.
5V Pranav, Assistant Professor, Department of Computer Science Engineering, Sir C R Reddy College of Engineering, India.

Manuscript received on 20 March 2019 | Revised Manuscript received on 25 March 2019 | Manuscript published on 30 July 2019 | PP: 6503-6507 | Volume-8 Issue-2, July 2019 | Retrieval Number: B2259078219/19©BEIESP | DOI: 10.35940/ijrte.B2259.078219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Reconstructing the damaged images and improving the quality of an image, results in image restoration. Here anisotropic diffusion based iterative inpainting developed to minimise the noise level in the colour images and enhancing the image boundaries, this approach observed on speckle, Gaussian and shot noise. To reduce noise and topological defects from images, 3D- anisotropic diffusion used to decompose the image into high frequencies and low frequencies and protects the image from losing the information, to enhance the image quality, image inpainiting was used. In this process most of the high frequency decomposed sections got damaged with noise and appears as there is information available at those pixels, therefore the complete restoration process was done on all the high frequency decomposed components so this results in achieving better restored images in mean time. The two effects on images can be reduced by the mixed fusion algorithm i.e., noise reduction by using anisotropic diffusion and distance-based neighbourhood image inpainting for restoring the damaged parts. So, this results in reconstructing the damaged image and enhancing the boundaries of the image.
Index Terms: Noise Enhancement, Image Restoration, Anisotropic Diffusion, Boundaries, Image Inpainting.

Scope of the Article: Image Processing and Pattern Recognition