Mirafs – Multi Image Resoulution Quality Acuminate Fusing System
John Livingston. S1, Kumudha Raimond2
1John Livingston. S, Assistant Professor, Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences. Coimbatore.
2Kumudha Raimond., Professor, Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences. Coimbatore.
Manuscript received on 15 August 2019 | Revised Manuscript received on 22 August 2019 | Manuscript published on 30 August 2019 | PP: 3031-3035 | Volume-8 Issue-10, August 2019 | Retrieval Number: J94690881019/19©BEIESP | DOI: 10.35940/ijitee.J9469.0881019
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 (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In Remote sensing applications, due to the increase in spatial and spectral resolutions, the time complexity became a vital aspect in order to process and deliver the solutions in a timely fashion. Remote sensors which acquire high-resolution image data often misses geometrical alignments and spectral stabilizations because of its dynamic routines, which results in crucial consideration such as loss of information and degraded quality in a single region of interest. This ignited us in finding the need for fusion of data to obtain quality products in remote sensing domain. Leaving the noise in source data compromises the overall quality of the solution. In proposed method to reduce the negative impacts of noise on the image skin, we use the confined filter function which held back the noise while reconstructing the image. In some cases, when image fusion is carried out using the least significant spectral band values will also lead to degraded output; in such condition, the Proposed method chooses high informative bands by careful attention to spectral values in each band. The anticipated outcomes through these methods are significant reduction in computational complexity and noise insensitivity. The fusion process must be verified to validate image sets for smart combinations of bands. The image pixels were considered as components to organize the fused spectral quality. The proposed HSI based MIRAFS method is compared with state of art fusion models and the results shows prominent improvement in the quality of fused images when we chose high spectral values. In MIRAFS, the quality of the resultant fused image has also been checked whether it has minimized spectral degradation with the amplified resolution by conserving spectral information as much as possible.
Index Terms: Fused Image; Noise; Spectral Quality; Image Pixel.
Scope of the Article: Image Security