Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/568042
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.coverage.spatial | ||
dc.date.accessioned | 2024-05-30T10:35:35Z | - |
dc.date.available | 2024-05-30T10:35:35Z | - |
dc.identifier.uri | http://hdl.handle.net/10603/568042 | - |
dc.description.abstract | Reinforcement Learning based real time object detection framework using deep machine learning model is a unique and important technique using which user can get quality output and can use in many system for getting efficient results. In this work we propose object detection with deep reinforcement learning by which we train the agent to extract the features of sequence of the frame and with a trained agent we detect the object present in video. The proposed technique improve the feature extraction ability of its convolution layers. Then, in the reinforcement learning based training phase, both the selection criteria for optimal action and the reward function are redesigned separately to explore more appropriate action and eliminate useless action .Finally, an effective online adaptive update strategy is proposed to adapt to the appearance changes or deformation of the object during actual tracking. Specially, meta-learning is utilized to pursue the most appropriate parameters for the network so that the parameters are closer to the optimal ones in the subsequent tracking process. The proposed tracker been improved from three aspects Firstly, the use of multi-domain training instead of supervised learning based training enables the tracker to learn the shared representation of different objects in the various training sequences. Secondly, the policy gradient based reinforcement learning is improved so that the tracker can capture the object by selecting more appropriate action and eliminating the useless action. Thirdly, the meta-learning based online adaptive update scheme is proposed to pursue the optimal parameters for the network. The proposed firstly video is divided into multiple frames {f_} ,these frames are the input frame {f_ . f_} ,each frame is combination of multiple patches {p_} then each frame goes to the convolutio layers ,convolution layers receive as input an image I(m and#8722; 1) and compute as output a new image I(m).Now depending upon the feature extraction tracker take the action {a_ . a__} in this | |
dc.format.extent | Xvii, 108 page | |
dc.language | English | |
dc.relation | ||
dc.rights | university | |
dc.title | Real Time Object Detection Framework Using Deep Machine Learning | |
dc.title.alternative | ||
dc.creator.researcher | Tiwari, Sonal | |
dc.subject.keyword | Computer Science | |
dc.subject.keyword | Computer Science Interdisciplinary Applications | |
dc.subject.keyword | Engineering and Technology | |
dc.subject.keyword | Meta-learning | |
dc.subject.keyword | Multi-domain training | |
dc.subject.keyword | Reinforcement Learning | |
dc.subject.keyword | Visual Tracing | |
dc.description.note | ||
dc.contributor.guide | Sharma, Shailja | |
dc.publisher.place | Bhopal | |
dc.publisher.university | Rabindranath Tagore University, Bhopal | |
dc.publisher.institution | Department of Computer Science and Engineering | |
dc.date.registered | 2020 | |
dc.date.completed | 2022 | |
dc.date.awarded | 2022 | |
dc.format.dimensions | ||
dc.format.accompanyingmaterial | DVD | |
dc.source.university | University | |
dc.type.degree | Ph.D. | |
Appears in Departments: | Department of Computer Science Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title page.pdf | Attached File | 540.14 kB | Adobe PDF | View/Open |
02_preliminary pages.pdf | 577.33 kB | Adobe PDF | View/Open | |
03_content.pdf | 54.7 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 84.73 kB | Adobe PDF | View/Open | |
05_chapter 1.pdf | 650.34 kB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 383.21 kB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 1.23 MB | Adobe PDF | View/Open | |
08_chapter 4.pdf | 4.42 MB | Adobe PDF | View/Open | |
09_chapter 5.pdf | 1.01 MB | Adobe PDF | View/Open | |
10_chapter 6.pdf | 382.54 kB | Adobe PDF | View/Open | |
11_chapter 7.pdf | 63.1 kB | Adobe PDF | View/Open | |
12_annexures.pdf | 15.33 MB | Adobe PDF | View/Open | |
80_recommendation.pdf | 524.43 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: