CBIR based on Colour, Edge Extraction Using Histogram and DWT

K. Ramanjaneyulu, Dr.K. Veera Swamy and Dr.Ch. Srinivasa Rao

This article discusses methods to extract images based on the colour and the edge feature extraction. In the suggested method, query image and image database, the feature vectors can be calculated using the colour and the edges using the histogram technique and discrete wavelet transform (DWT). The similarity between image and query image is attained by computing the variance between their feature vectors with the help of distance measure. Vectors of the images with smaller distance are the most identified with the query image. The robustness of the suggested technique is also examined the variations of the query image, such as transformation, rotation and adding noise or ambiguity. Corel image database is taken for experimental study and system performance is calculated by precision and recall.

Volume 11 | 07-Special Issue

Pages: 1618-1625