Autor: |
Sandip M. Walunj, Amol D. Potgantwar, Shivani V. Gaikar |
Rok vydání: |
2017 |
Předmět: |
|
Zdroj: |
2017 International Conference on Computing, Communication, Control and Automation (ICCUBEA). |
DOI: |
10.1109/iccubea.2017.8463862 |
Popis: |
The Region-of-Interest methodologies of image processing allocate more bits to ROI's and less bits to different areas. The perceptual character of image reduces as the background image gets deformed. To address the issue, the framework produces hierarchical system of 3 layers i.e. image search, patch matching and image synthesis(image reconstruction). The Proposed system is made to work in serial as well as in parallel. On comparing the results of serial and parallel, better results are achieved by using parallel way in terms of Time, Recall, Precision and Fmeasure(Accuracy). The proposed system enhances the speed of reconstruction of reconstruction of image by 5-6s% and also increases the accuracy by 8% to 10% than existing approach. |
Databáze: |
OpenAIRE |
Externí odkaz: |
|