A Hybrid Approach to Feature Extraction for Content Based Image Retrieval
Er. Shilpa, Er. Navdeep Singh · International Journal of Advanced Research in Computer Science · 2016
Query image retrieval which is also known as Content Based Image Retrieval (CBIR) is defined as the retrieving of the images from the database that have content similar to the query image.It is done by extracting features according to the content in the form of colour, shape and texture from the images. There exists various techniques for the retrieval of images from the database. Though these techniques work well for retrieving the images from the database, still they lack in accuracy. In this paper, a hybrid technique is proposed for efficient retrieval of images. Rotation Invariant Uniform LBP (RIULBP) with colour moment and Centre-Symmetric LBP (CSLBP) with colour moment together forms hybrid approach. The proposed technique has an accuracy of 86.1% which is fairly high in comparison to if these techniques are used in isolation. Keywords-Content Based Image Retrieval, LBP operator, Uniform local pattern operator. Rotation Invariance, Centre-Symmetric LBP.