Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/72116
Title: Mobile visual product search and recommendation
Authors: Aung, Thin Thin
Keywords: DRNTU::Engineering::Electrical and electronic engineering
Issue Date: 2017
Abstract: Nowadays, computer vision is one of the popular developing fields which gives more effective data tracking and tracing services to the users. Hence, visual recognition based applications for the smartphone have drawn grown interest and demand by users lately. Although there are many studies of visual recognition based applications in various areas, the study of visual recognition based application for artifacts is still not widely covered and developed yet. Therefore, this project mainly discusses the insight of visual recognition techniques and conducts experiments to observe the accuracy rate of the technique on artifacts. During this project, a small database with minimum criteria set is created and used for experiments. Moreover, Geometric Verification (GV), an alternative method which can reduce external noises such as illuminations changes, orientation changes and occlusions is introduced. After several experiments, results showed an increase of matching accuracy between 5 to 10% by this method. However, one of the disadvantages of this method is it required longer processing time for feature matching step. Lastly, various sets of experiments with different conditions are conducted with and without GV. Then, result analysis of the experiments are made and compared in term of processing time and accuracy rate. In conclusion, the possible ways to enhance the matching efficiency as well as to further test algorithm robustness for the future studies are discussed.
URI: http://hdl.handle.net/10356/72116
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
THINTHINAUNG_U1322922K_FINALREPORT.pdf
  Restricted Access
3.12 MBAdobe PDFView/Open

Page view(s)

126
Updated on May 16, 2021

Download(s)

14
Updated on May 16, 2021

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.