Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/253350
Title: | Studies on visual saliency detection models and their applications in image segmentation and image compression |
Researcher: | Diana andrushia A |
Guide(s): | Thangarajan R |
Keywords: | Engineering and Technology,Computer Science,Computer Science Information Systems Image segmentation visual saliency |
University: | Anna University |
Completed Date: | 2018 |
Abstract: | Humans can be aware of or understand the environment they live in newlinefrom different levels of observation and participation. This is mainly due to newlinethe innate ability of the human brain in processing visual information by newlinevirtue of abundance of data observed and accumulated over years of newlineexistence. Saliency detection in a visual field is another unparalleled ability of newlinethe human brain. The brain without any conscious effort immediately draws newlineits attention to any salient of prominent features present in an image or scene. newlineWhen the ability of automatically identifying the important regions of an newlineimage or scene, it would serve as an indispensable tool for many applications newlinein the domain of computer vision, image analysis and artificial intelligence. newlineThis research work mainly focuses on visual information processing of images or scenes with regard to saliency detection. Visual scenes may represent various aspects of human day to day affairs such as newlinemundane tasks, arts, holiday outs. Important or salient content of an image newlinegenerally stands out from the rest of the image. Obviously, analysis of images newlinedemands high computational power, abundance of data, and efficient learning newlinealgorithms. Feature engineering is also required to aid machine learning newlinealgorithms to identify which part of an image contains important information newlinewith respect to the rest other parts. This means lesser amount of data, simpler newlinemodels, and more accuracy with lower variance.. newline newline |
Pagination: | xx, 145p. |
URI: | http://hdl.handle.net/10603/253350 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 24.19 kB | Adobe PDF | View/Open |
02_certificates.pdf | 1.02 MB | Adobe PDF | View/Open | |
03_abstract.pdf | 13.13 kB | Adobe PDF | View/Open | |
04_acknowledgment.pdf | 4.6 kB | Adobe PDF | View/Open | |
05_contents.pdf | 33.06 kB | Adobe PDF | View/Open | |
06_chapter1.pdf | 374.99 kB | Adobe PDF | View/Open | |
07_chapter2.pdf | 243.41 kB | Adobe PDF | View/Open | |
08_chapter3.pdf | 846.66 kB | Adobe PDF | View/Open | |
09_chapter4.pdf | 1.58 MB | Adobe PDF | View/Open | |
10_chapter5.pdf | 708.58 kB | Adobe PDF | View/Open | |
11_chapter6.pdf | 588.96 kB | Adobe PDF | View/Open | |
12_conclusion.pdf | 29.03 kB | Adobe PDF | View/Open | |
13_references.pdf | 170.79 kB | Adobe PDF | View/Open | |
14_publications.pdf | 127.81 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: