Abstract—Large image databases find diverse applications in real-life situations. It is essential to develop an efficient technique to grasp required information from these databases. A good number of researchers are involved in developing techniques for object retrieval and recognition using different image features, such as color histogram, distance, and shape parameters. Efficient retrieval and robustness are two main criteria for recognition of an object using image databases. Keeping this in mind, the present paper describes a hybrid method using a weighted combination of color histogram, distance, and moment parameters for trademark recognition. Our experimentation with a database containing 200 trademark images confirms the superiority of the present method in comparison to the techniques with single feature.
Index Terms—Color histogram, distance parameter, image recognition, moment invariant, trademark.
M. Sultana is with the Department of Computer Science and Engineering, University of Liberal Arts Bangladesh, Dhaka, Bangladesh (e-mail: madeena.sultana@ ulab.edu.bd).
M. S. Uddin is with the Department of Computer Science and Engineering, Jahangirnagar University, Dhaka, Bangladesh (e-mail:firstname.lastname@example.org).
Cite: Madeena Sultana and Mohammad Shorif Uddin, "Trademark Recognition Using a Weighted Combination of Different Image Features," International Journal of Computer Theory and Engineering vol. 4, no. 6, pp. 1035-1038, 2012.