Content-Based Image Retrieval using Features Extraction and Rotation Invariant Feature Transformation

Content-based image retrieval uses the visual con-tents of an image such as color, shape, texture to represent and index the image. In this system the visual contents of the images in the database are extracted and described by features. The features of the images in the database form a feature database. Feature database is used for similarity comparison with query image. Then CBIR system gives images relevant to query image from dataset images as a output. Content-based image retrieval (CBIR) by extracting the advantage of low complexity ordered dither block truncation coding (ODBTC) for the generation of image content descriptor. In encoding, ordered dither block truncation coding (ODBTC)[1] compresses an image block into corresponding quantizers and bitmap image. We proposed two image features to index an image are color co-occurrence feature (CCF) and bit pattern features (BPF), which are generated directly from the ordered dither block truncation coding (ODBTC). The color co-occurrence feature (CCF) and bit pattern features (BPF) of an image are simply derived from the two ODBTC quantizers and bitmap, respectively, by involving the visual codebook. We have also proposed a novel feature representation method for content-based image retrieval is rotation-Invariant for checking retrieval accuracy of the proposed method depends on its rotation invariant ability or not.