Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/331502
Title: | Traffic sign detection using intelligent classifiers for enhanced driver guidance system |
Researcher: | Jayaprakash A |
Guide(s): | Kezi selvavijila C |
Keywords: | Engineering and Technology Engineering Engineering Electrical and Electronic driver guidance system Traffic sign |
University: | Anna University |
Completed Date: | 2020 |
Abstract: | Road Sign Detection and Recognition (RSDR) due to its challenging nature as a computer vision problem has become a hot area of research in the recent years. Road traffic signs provide instructions and cautioning information in order to regulate the driver behaviour. RSDR systems helps in detecting and translating the road signs forthe attention and understanding of drivers, in combination with other Advanced Driver Assistance Systems (ADAS) such as lane departure warning systems, in-car navigation systems, adaptive cruise control system, automatic parking etc. Particularly, it serves to be a great help to the visually impaired driverswho benefit from this important computer based visual aid. In more classy systems, RSDR systems can utilise other features of ADAS such as adaptive cruise control system which helps in the automatic drive of the vehicle in accordance with the varying road speeds. In this thesis, Histogram Orientation Gradients (HOG) and Support Vector Machines (SVMs) are chosen to build the RSDR system. For better results, different HOG parameters have been tested to find the best combination. Maximally Stable Extremal Region (MSER) is found to have reduced time consumption compared to HOG. The second part of the RSDR system based on Content Recognition, performed by extracting the Local Energybased Shape Histogram (LESH), features the normalized road sign contents. The extracted content features are utilised to train a Support Vector Machine (SVM) polynomialkernel toperform the classification precisely. Finally the thesis also explains a new approach which comprises preprocessing, edge detection, feature extraction, features selection by using Ensemble Fuzzy Support Vector Machine (EFSVM) classifier. Feature selection is carried out successfully by the deployment of Ant Colony Optimization (ACO) algorithm to determine most prominent and definitive features newline |
Pagination: | xvii, 150p. |
URI: | http://hdl.handle.net/10603/331502 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 226.22 kB | Adobe PDF | View/Open |
02_certificates.pdf | 337.56 kB | Adobe PDF | View/Open | |
03_vivaproceedings.pdf | 6.85 MB | Adobe PDF | View/Open | |
04_bonafidecertificate.pdf | 12.91 MB | Adobe PDF | View/Open | |
05_abstracts.pdf | 53.87 kB | Adobe PDF | View/Open | |
06_acknowledgements.pdf | 993.23 kB | Adobe PDF | View/Open | |
07_contents.pdf | 72.72 kB | Adobe PDF | View/Open | |
08_listoftables.pdf | 53.31 kB | Adobe PDF | View/Open | |
09_listoffigures.pdf | 68.33 kB | Adobe PDF | View/Open | |
10_listofabbreviations.pdf | 82.95 kB | Adobe PDF | View/Open | |
11_chapter1.pdf | 2.81 MB | Adobe PDF | View/Open | |
12_chapter2.pdf | 1.9 MB | Adobe PDF | View/Open | |
13_chapter3.pdf | 188.04 kB | Adobe PDF | View/Open | |
14_chapter4.pdf | 9.59 MB | Adobe PDF | View/Open | |
15_chapter5.pdf | 6.43 MB | Adobe PDF | View/Open | |
16_chapter6.pdf | 4.64 MB | Adobe PDF | View/Open | |
17_conclusion.pdf | 151.7 kB | Adobe PDF | View/Open | |
18_appendices.pdf | 8.18 MB | Adobe PDF | View/Open | |
19_references.pdf | 214.81 kB | Adobe PDF | View/Open | |
20_listofpublications.pdf | 148.84 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 169.52 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: