Content based image retrieval using features extracted from halftoning-based block truncation coding
Download as DOCX, PDF0 likes466 views
The document presents a technique for content-based image retrieval (CBIR) using low-complexity ordered-dither block truncation coding (ODBTC) to create image content descriptors. It introduces two features, color co-occurrence feature (CCF) and bit pattern features (BPF), which can be derived without decoding the image. Experimental results demonstrate the superiority of this method over previous image retrieval systems, highlighting ODBTC's dual functionality in compression and effective image indexing.
1 of 1
Download to read offline
More Related Content
What's hot (18)
Ad
Similar to Content based image retrieval using features extracted from halftoning-based block truncation coding (20)
Ad
More from I3E Technologies (20)
Recently uploaded (20)
Content based image retrieval using features extracted from halftoning-based block truncation coding
- 1. CONTENT-BASED IMAGE RETRIEVAL USING FEATURES EXTRACTED FROM HALFTONING-BASED BLOCK TRUNCATION CODING ABSTRACT A technique for content-basedimage retrieval (CBIR) by exploiting the advantage of lowcomplexityordered-dither block truncation coding (ODBTC) forthe generation of image content descriptor. In the encodingstep, ODBTC compresses an image block into corresponding quantizes and bitmap image. Two image features are proposedto index an image, namely, color co-occurrence feature (CCF) andbit pattern features (BPF), which are generated directly from theODBTC encoded data streams without performing the decodingprocess. The CCF and BPF of an image are simply derived from the two ODBTC quantizes and bitmap, respectively, by involvingthe visual codebook. Experimental results show that the proposedmethod is superior to the block truncation coding image retrievalsystems and the other earlier methods, and thus prove that theODBTC scheme is not only suited for image compression, becauseof its simplicity, but also offers a simple and effective descriptorto index images in CBIR system.