Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/368044
Title: | Semantic Based Video Searching for Organizing Multimedia Big Data |
Researcher: | A Manju |
Guide(s): | Valarmathie P |
Keywords: | Automation and Control Systems Computer Science Engineering and Technology |
University: | Saveetha University |
Completed Date: | 2021 |
Abstract: | With the proliferation of the internet, big data continues to grow exponentially, newlineand video has become the largest source. Video big data introduces many newlinetechnological challenges, including compression, storage, transmission, analysis, and newlinerecognition. The increase in number of multimedia resources has brought an urgent newlineneed to develop intelligent methods to organize and process them. The integration newlinebetween Semantic link Network and multimedia resources provides a new prospect newlinefor organizing them with their semantics. The tags and surrounding texts of multimedia newlineresources are used to measure their semantic association. Two evaluation methods newlineincluding clustering and retrieval are performed to measure the semantic relatedness newlinebetween images accurately and robustly. This method is effective on image searching newlinetask. newlineFirst a Video Semantic Content Extraction Framework is presented which uses newlineTensor Flow. The method read the video files and generates set of video frames. newlineFrom each frame, the method applies Gabor filter and eliminates the noise from the newlineframes. Then, the quality of frame is improved by applying the histogram equalization newlinetechnique. Further, the method eliminates the background and identifies the object newlinefrom the frames. Using the objects, the method extracts spatial, temporal and newlineconceptual features. Such features extracted are used to classify the video file. newlineThe usage of semantics can be explored for video searching. A whole model newlinefor generating the association relationship between video resources using Semantic newlineLink Network model is proposed. The user can select the defined attributes or concepts newlineas the searching queries. This can be done by providing the knowledge conduction newlineduring information extraction and by applying fuzzy reasoning. First a video semantic newlinecontent extraction framework was proposed to track the object. This framework uses newlineGabor filter to eliminate noise and quality was improved by histogram equalization newlinetechnique. |
Pagination: | |
URI: | http://hdl.handle.net/10603/368044 |
Appears in Departments: | Department of Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf.pdf | Attached File | 49.52 kB | Adobe PDF | View/Open |
02_certificate.pdf.pdf | 77.4 kB | Adobe PDF | View/Open | |
03_abstract.pdf.pdf | 26.93 kB | Adobe PDF | View/Open | |
04_declaration.pdf.pdf | 36 kB | Adobe PDF | View/Open | |
05_acknowledgement.pdf.pdf | 77.5 kB | Adobe PDF | View/Open | |
06_contents.pdf.pdf | 34.49 kB | Adobe PDF | View/Open | |
07_list_of_tables.pdf.pdf | 27.29 kB | Adobe PDF | View/Open | |
08_list_of_figures.pdf.pdf | 82.52 kB | Adobe PDF | View/Open | |
09_abbreviations.pdf.pdf | 25.56 kB | Adobe PDF | View/Open | |
10_chapter1.pdf.pdf | 327.85 kB | Adobe PDF | View/Open | |
11_chapter2.pdf.pdf | 150.66 kB | Adobe PDF | View/Open | |
12_chapter3.pdf.pdf | 472.43 kB | Adobe PDF | View/Open | |
13_chapter4.pdf.pdf | 361.92 kB | Adobe PDF | View/Open | |
14_chapter5.pdf.pdf | 631.3 kB | Adobe PDF | View/Open | |
15_chapter6.pdf.pdf | 198 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 146.8 kB | Adobe PDF | View/Open | |
bibliography.pdf | 113.14 kB | Adobe PDF | View/Open | |
conclusion and summary.pdf | 146.8 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: