Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/195657
Title: | Certain Investigations on Gait Recognition using Efficient Data Mining Algorithms |
Researcher: | Aasha M |
Guide(s): | Sivakumari S |
Keywords: | Data Mining Gait Recognition Feature Extraction |
University: | Avinashilingam Deemed University For Women |
Completed Date: | 28.07.2017 |
Abstract: | Gait recognition system is becoming an increasingly important means for identification of humans in newlinethe present world. Gait recognition identifies people by the way they walk and are utilized in different security newlinesensitive application areas such as banks, large-scale industries etc. Gait identification task becomes difficult newlinedue to the change of appearance by different cofactors (e.g., shoe, surface, carrying, view, and clothing).In newlinerecent years more emphasis was given to part based gait recognition techniques which segment the human body newlineinto different parts based on the effect of different cofactors. Some parts of gait are affected by cofactors and newlineother parts remains unaffected. Initially in this work, adaptive fusion of part based gait identification is newlineproposed. The proposed gait identification adaptively fuses the best informative less effective part with the most newlineeffective parts. Feature extraction is an important step in gait recognition and therefore to improve the newlineefficiency of the system, Multi objective Bat algorithm is proposed in which the shape descriptor features are newlineincluded to improve the accuracy of gait recognition. newlineFurthermore to reduce the search space time, gender classification has been introduced which reduces newlinethe total number of search subjects from database. Gender classification is performed by utilizing Sparse newlinespatiotemporal features along with most effective features, more informative less effective features and shape newlinefeatures. In order to further improve the accuracy of the system, the velocity and depth features of gait are newlineextracted along with most effective and less effective features and shape features. Here, velocity moments and newlinedepth are measured by velocity measurement and feature vector calculation method respectively. newlineThe system is further enhanced by using an efficient gait recognition using multi-objective effective newlineenhanced adaptive fusion technique by hybrid MPSO-BAT algorithm. |
Pagination: | 102 p. |
URI: | http://hdl.handle.net/10603/195657 |
Appears in Departments: | Department of Computer Science and Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 758.64 kB | Adobe PDF | View/Open |
02_certificate.pdf | 759.8 kB | Adobe PDF | View/Open | |
03_acknowledgement.pdf | 795.35 kB | Adobe PDF | View/Open | |
04_contents.pdf | 795.38 kB | Adobe PDF | View/Open | |
05_list of tables, figures and abbreviations.pdf | 793.47 kB | Adobe PDF | View/Open | |
06_chapter 1.pdf | 7.62 MB | Adobe PDF | View/Open | |
07_chapter 2.pdf | 7.53 MB | Adobe PDF | View/Open | |
08_chapter 3.pdf | 7.46 MB | Adobe PDF | View/Open | |
09_chapter 4.pdf | 7.69 MB | Adobe PDF | View/Open | |
10_chapter 5.pdf | 7.61 MB | Adobe PDF | View/Open | |
11_chapter 6.pdf | 7.65 MB | Adobe PDF | View/Open | |
12_chapter 7.pdf | 7.59 MB | Adobe PDF | View/Open | |
13_chapter 8.pdf | 7.58 MB | Adobe PDF | View/Open | |
14_chapter 9.pdf | 7.51 MB | Adobe PDF | View/Open | |
15_chapter 10.pdf | 7.41 MB | Adobe PDF | View/Open | |
16_references.pdf | 7.51 MB | 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: