Exploiting image indexing techniques in DCT domain

Abstract

This paper is concerned with the indexing and retrieval of images based on features extracted directly from the JPEG discrete cosine transform (DCT) domain. We examine possible ways of manipulating DCT coe$cients by standard image analysis approaches to describe image shape, texture, and color. Through the Mandala transformation, our approach groups a subset of DCT coe$cients to form ten blocks. Each block represents a particular frequency content of the original image. Two blocks are used to model rough object shape; nine blocks to describe subband properties; and one block to compute color distribution. As a result, the amount of data used for processing and analysis is signi"cantly reduced. This can lead to simple yet e$cient ways of indexing and retrieval in a large-scale image database. Experimental results show that our proposed approach o!ers superior indexing speed without signi"cantly sacri"cing the retrieval accuracy.

Block selective redaction for minimizing loss during de-identification of burned in text in irreversibly compressed JPEG medical images

“…[21][22][23][24][25][26][27][28][29] Selective use of blocks in the transformed domain for other purposes, such as searching, indexing, or selective region decompression has also been described. [30][31][32][33] Pure Java frameworks are popular for deidentification, whether they use an automated bulk process controlled by templates, such as RSNAs Clinical Trial Processor (CTP), 34 or an interactive user-controlled process, with a tool such as DicomCleaner. 35 Accordingly, it was necessary to implement a Pure Java block selective redaction process rather than to modify existing platform-specific code.…”

Image Similarity based on a Distributional "Metric" for Multivariate Data

“…To bridge the gap between compressed-and pixel-space, where the majority of image processing algorithms are developed, recent research is now starting apace to develop content feature extraction algorithms working directly in the compressed domain (e.g. Zhong & Jain, 2000;Ngo et al, 2001;Jiang et al, 2004). Since the inverse DCT (IDCT) is an embedded part of the JPEG decoder and the DCT itself is one of the best filters for feature extraction working directly on the DCT domain, it has proven to be a well-promising area for image similarity in the compressed domain.…”

“…Since the inverse DCT (IDCT) is an embedded part of the JPEG decoder and the DCT itself is one of the best filters for feature extraction working directly on the DCT domain, it has proven to be a well-promising area for image similarity in the compressed domain. DCT has, to a certain extent, unique scale invariance and zooming characteristics, which can provide insight into objects and texture identification (Ngo et al, 2001). In addition, it exhibits a set of good properties such as energy compaction and image data decorrelation and, therefore, is naturally considered to be a potential domain in mining visual information.…”

“…An attempt is described here to adopt our methodology so as to work in compressed image domains that have recently gained high popularity (e.g. Zhong & Jain, 2000;Ngo et al, 2001;Jiang et al, 2004). This is expected not only to increase the efficiency of WW-based similarity scheme -by avoiding image decompression -, but also to constitute it suitable for novel applications like searching and retrieval in the World-Wide-Web -since the images of the Web are mostly included in a standard compressed format - (Jiang et al, 2004).…”

Retrieval of Images Using DCT and DCT Wavelet Over Image Blocks

“…These attributes can be used and processed to represent the image feature to make them comparable for similarity. Many techniques are being developed in this field to retrieve the images from large volume of database more precisely [1], [2], [3], [11], [12], [13] [32] [33]. This paper contributes in same direction by introducing the novel techniques which are giving favorable performance which is analyzed through different aspects of the behavior of the proposed CBIR system.…”