Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/314899
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.coverage.spatial | ||
dc.date.accessioned | 2021-02-12T07:07:24Z | - |
dc.date.available | 2021-02-12T07:07:24Z | - |
dc.identifier.uri | http://hdl.handle.net/10603/314899 | - |
dc.description.abstract | Presently, visual tracking become more popular in the area of computer vision. newlineFor effective object tracking, the tracking method should have the capability to newlineseparate the target objects from background accurately. While designing the model of newlinevisual tracking, several issues need to be considered. Some of the issues are occlusion, newlinescale variation, rotation, motion blur, deformation and background clutter. In order to newlineachieve effective visual tracking, numerous visual tracking methods have been newlinedeveloped. Existing tracking methods works well and produce effective results only in newlinesimpler situations, i.e. slow motion, less occlusion, etc. Tracking generic objects is still newlinea difficult due to the presence of blur, rotation, fast motion, occlusion, scale variation, newlinepose change and background noise. Though single tracking models are newlinecomputationally less complex, it is not suitable for situations with the existence of newlineocclusion and noise. Multi-Object Tracking (MOT) approaches manage occlusion in newlinean effective way by the use of high level associations. MOT methods are newlinecomputationally more complex than single tracking method. For visually tracking the newlineobject in complicated situations, in recent days, Deep Learning (DL) models are also newlinedeveloped for detecting multiple objects. The existing tracking methods fail to produce newlineeffective results in complicated situations. To resolve these issues, in this research newlinework, we focus on the design on the multi-object detection and tracking mode | |
dc.format.extent | ||
dc.language | English | |
dc.relation | ||
dc.rights | university | |
dc.title | a study and development of robust multi object recognition and tracking model using mask region based convolution neural network | |
dc.title.alternative | ||
dc.creator.researcher | A. Nirmala | |
dc.subject.keyword | Engineering and Technology | |
dc.description.note | ||
dc.contributor.guide | S. Arivalagan | |
dc.publisher.place | Annamalai Nagar | |
dc.publisher.university | Annamalai University | |
dc.publisher.institution | Department of Computer and Information Science | |
dc.date.registered | 2017 | |
dc.date.completed | 2020 | |
dc.date.awarded | ||
dc.format.dimensions | ||
dc.format.accompanyingmaterial | DVD | |
dc.source.university | University | |
dc.type.degree | Ph.D. | |
Appears in Departments: | Department of Computer and Information Science |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
10. chapter 3.pdf | Attached File | 631.5 kB | Adobe PDF | View/Open |
11. chapter 4.pdf | 2.14 MB | Adobe PDF | View/Open | |
12. chapter 5.pdf | 266.21 kB | Adobe PDF | View/Open | |
13. chapter 6.pdf | 469.08 kB | Adobe PDF | View/Open | |
14. chapter 7.pdf | 30.57 kB | Adobe PDF | View/Open | |
1. cover page.pdf | 356.89 kB | Adobe PDF | View/Open | |
2. certificate.pdf | 376.59 kB | Adobe PDF | View/Open | |
4. acknowledgement.pdf | 146.18 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 30.57 kB | Adobe PDF | View/Open | |
8. chapter 1.pdf | 835.37 kB | Adobe PDF | View/Open | |
9. chapter 2.pdf | 474.9 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: