Abstract—This paper proposes a content-based image retrieval system based on an efficient combination of both feature and color algorithms. According to Kekre Transform we derive feature vectors using a combination of row mean and column mean of both query as well as the database images, to measure the extent of similarity in features using Euclidian distance. Experimentation shows that taking row mean, column mean and combination improves the performance of image retrieval as compared to taking Kekre Transform of the whole image. Similarly, HSV colour space quantifies the colour space into different regions and thereby calculating its mean and Euclidian distance the colour vector can be derived. Taking mean or Euclidian distance of the individual Euclidian distances of both the algorithms improves the accuracy of the image retrieval process implemented. By calculating the precision and recall parameters of selective images from the database, comparison between the two algorithms and the effectiveness of the combination of both the algorithms can be measured.
Index Terms—CBIR, Feature Vector, Color Vector, Euclidian distance, HSV, Hue, Adaptive Segmentation
Venu Shah is a Sr. Professor with Atharva College of Engineering, Malad (W), Mumbai-400095, India (email: email@example.com).
Pavan Bhat, Mahesh Dasarath, Shipra Gupta, Jackie Bhavsar are B.EEXTC students of Atharva College of Engineering, Malad (W), Mumbai-400095, India.(email: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com)
Cite: Venu Shah, Pavan Bhat, Mahesh Dasarath, Shipra Gupta and Jackie Bhavsar, "Content Based Image Retrieval Using Combination of Kekre Transform and HSV Color Segmentation," International Journal of Engineering and Technology vol. 3, no. 5, pp. 547-552, 2011.