High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Content Based Trademarks Retrieval using Bag of Visual Words

Author(s):

M.Ramesh Kumari , Ayyanadar janaki ammal polyechnic college, Sivakasi

Keywords:

Multimedia Object, Text Detection, Localization, Tracking, Extraction, Enhancement

Abstract

Trademarks are distinctive visual symbols used to identify and distinguish a specific product, service or business. With the rapid increase in the amount of registered trademark images around the world, Trademark Image Retrieval (TIR) has emerged to ensure that new trademarks do not repeat any of the vast number of trademark images stored in the trademark registration system. Content Based Image Retrieval (CBIR) method is a promising method for performing trade mark retrieval. In this work, Histogram of Oriented radiant (Hog) algorithm is developed to draw the histogram for the gradient image for the given trade mark logo. Addition, an L1 Distance algorithm is developed to find similarity between query image and database images by comparing the corresponding histograms and to obtain images similar to query image from data base. Precision and recall method is used to evaluate the performance of above mentioned algorithms and results obtained are compared with the results of SIFT algorithm obtained from literature. It is found that the HOG algorithm is better with 96% of retrieval efficiency than SIFT algorithm which has retrieval efficiency of 94% only.

Other Details

Paper ID: IJSRDV6I60108
Published in: Volume : 6, Issue : 6
Publication Date: 01/09/2018
Page(s): 190-195

Article Preview

Download Article