Image indexing and retrieval based on vector quantization

Indexing chromatic and achromatic patterns for content-based colour image retrieval

A critical evaluation of image and video indexing techniques in the compressed domain

The theme of the work presented here is image retrieval using texture patterns generated from Haar transform matrix and image bitmaps. Different texture patterns namely '4-pattern', '16-pattern', '64-pattern', '256-pattern' and '1024-pattern' are generated using Haar transform matrix and then compared with the bitmap of an image to generate the feature vector as the matching number of ones and minus ones per texture pattern. The proposed content based image retrieval (CBIR) techniques are tested on a generic image database having 1000 images spread across 11 categories. For each proposed CBIR technique 55 queries (randomly selected 5 per category) are fired on the image database. To compare the performance of image retrieval techniques, crossover point of average precision and recall values of all the queries are computed per image retrieval technique. Ameliorated performance (higher precision and recall values) has been observed with the proposed methods compared to the colour averaging based image retrieval techniques. Further the performance of proposed image retrieval methods is enhanced using the combination of original image and even image part. In the discussed image retrieval methods, the combination of original and even image part for 256-pattern texture gives the highest crossover point of precision and recall reflecting better performance.

The spatial relation model is an important technique for image indexing and retrieval in image or multimedia databases. 2D strings and its variants are proposed to support the representation of spatial relationships. A novel spatial knowledge representation model named “Two Dimension Begin-End boundary string” (2D Bε-string) is proposed. The 2D Bε-string represents an icon by its MBR boundaries. By applying a number of “dummy objects”, the 2D Bε-string can intuitively and naturally represent the pictorial spatial information without any spatial operator. In addition, an image similarity evaluation method based on the modified “Longest Common Subsequence” (LCS) algorithm is presented. By using the proposed evaluation method, not only can those images for which all the icons and their spatial relationships fully agree with the query image be sifted out, but also those images whose partial icons and/or spatial relationships are similar to the query image. It resolves the problems of query targets and/or spatial relationships not being certain. Our representation model and similarity evaluation also simplify the retrieval progress of linear transformations, including image rotation and reflection

The vast growth of Internet and digital contents of Internet creates much interest on automated image indexing and retrieval techniques. Most commercial Content Based Image Retrieval systems use color histograms as one of the features. In this work, we have prepared combined color histograms for three regions (top, bottom and middle) of the image. The combination of those histograms will indirectly include the space information within it. Besides this, straight-line signatures and outline sketch signatures from the image are extracted. The average precision is used as a metric to evaluate the performance of the system. Experiments were successfully implemented on 1000 images of SIMPLIcity database using MAT Lab. The performance of our method is compared with four other previous methods Integrated Region Matching[l], Fuzzy Integrated Region Matching [2], Color Salient Points and shape features [3] and Interactive CBIR using Ripplet Transform and Fuzzy Relevance Feedback Mechanism[4] which used the same data set and evaluation method. According to the arrived results, the performance of the proposed Three Region Color, Straight-Line and Outline Sketch Signatures based CBIR system was significantly better than other compared methods.

An ideal image compression technique for image data archival, retrieval and distribution would be one with the asymmetrical computational requirements of vector quantization (VQ), but without the complications arising from VQ codebooks. Codebook generation and maintenance are stumbling blocks which have limited the use of VQ as a practical image compression algorithm. Model-based VQ (MVQ), a variant of VQ described here, has the computational properties of VQ but does not require explicit codebooks. The codebooks are internally generated using mean removed error and human visual system (HVS) models. The error model assumed is the Laplacian distribution with mean, (lambda) , computed from a sample of the input image. A Laplacian distribution with mean, (lambda) , is generated with a uniform random number generator. These random numbers are grouped into vectors. These vectors are further conditioned to make them perceptually meaningful by filtering the DCT coefficients from each vector. The DCT coefficients are filtered by multiplying by a weight matrix that is found to be optimal for human perception. The inverse DCT is performed to produced the conditioned vectors for the codebook. The only image dependent parameter used in the generation of codebook is the mean, (lambda) , that is included in the coded file to repeat the codebook generation process for decoding.

International audience

Image retrieval is a critical task in many fields such as indexing and Content-Based Image Retrieval is a tried and tested method for image retrieval. With advances in graphics and visualization techniques, image retrieval should be easy, quick and flexible. The aim of this work is to develop an efficient image indexing and retrieval system to collate similar images or images based on conditioned queries in the form of an image itself. The indexing method has been performed using wavelet transform and related mechanisms. Its detailing can easily entail description from a broad range to a narrow range. The indexing algorithm applies a Haar Wavelet-Based Perceptual Similarity Index (HaarPSI) for Image Quality Assessment. The HaarPSI utilizes the coefficients obtained from a Haar wavelet decomposition to assess local similarities between two images, as well as the relative importance of image areas. Results show that the algorithm used is able to retrieve and display images with high retrieval accuracy and medium relevance.The proposed methodology has been tested against known image retrieval techniques and was found to be comparable and in some cases better than most prior known techniques. The testing was performed on images of varied types, from multiple datasets and sizes. The accuracy was found to be 99% in most cases. All in all, this study introduces a simple and convenient way of offline image searches on desktop computers and provides a stepping stone to future content-based image retrieval systems built for similar purposes.

The proliferation of computer technology and digital image‐acquisition hardware has led to the widespread use of image data across a variety of applications including astronomy, art, natural resources, engineering design, military, business operation, medicine, education, etc. Major research activities in the digital image databases surged after the U.S. government's Digital Library Initiative (DLI) from 1994 to 1998. The DLI research team at the University of California at Berkeley developed a work‐centered digital information system that contains 450 digital text documents, 200 air photos, and over 11,000 ground photographs. The system allowed a user or a working group to access its own collections of varying data types, and to generate new materials to be added to the collection. The DLI project at the University of California (UC) at Santa Barbara focused on the development of digital databases that contain geographically referenced materials such as maps and aerial photos. The research team developed a number of tools for browsing and retrieving map images at multiple resolution. Two other major efforts in developing digital image databases were at The National Library of Medicine (NLM), a component of the National Institutes of Health (NIH), and at Time Warner. More recently, research scientists at the University of California at Berkeley and the Fine Arts Museums of San Francisco successfully launched on the Internet the largest art image database in the world, the Thinker ImageBase.Automatic image indexing and retrieving technology is fundamental in digital image databases. There are two general strategies in image retrieval, browsing and searching. Browsing is an information retrieval strategy in which a user navigates through an ordered arrangement of images by making selections from the progressive levels of a hierarchy into which the available images have been logically grouped. Image browsing relies on the user's cognitive abilities to recognize images of interest without having to formulate a specific query. Searching is an information retrieval strategy in which the user communicates with an information retrieval system by an interface that requires the user to input a search query.Images can be indexed and retrieved by text‐tag information and/or by image content. Conventionally, images are indexed using text‐tag information, such as title, key words, date of the work, artist, author, photographer, legend, captions, etc. Images are often subject to a wide range of interpretations, and textual descriptions can only begin to capture the richness and complexity of the semantic content of a visual image. In many applications, the complexity of the information embedded in image content such as types of objects, object attributes, and spatial relationship of objects, etc, cannot be synthesized in a few key words. It has been reported that users querying an image collection tend to be much more specific in their requests and information needs than when querying a text database. Furthermore, text indexing requires large human effort in creating the meta‐data that enables visual queries and is language dependent. Large databases containing thousands and millions of digital images that can occupy gigabytes of space are almost impossible for manual indexing and searching. The demand for systems using pictorial information combined with textual description in image retrieval is growing and automatic indexing and retrieval based on image content has become the most promising techniques for large image databases and digital image libraries. This article describes the techniques developed for content‐based retrieval.

Multi-joint histogram based modelling for image indexing and retrieval

With the rapid growing volume of images in medical databases, development of efficient image retrieval systems to retrieve relevant or similar images to a query image has become an active research area. Despite many efforts to improve the performance of techniques for accurate image retrieval, its success in biomedicine thus far has been quite limited. This article presents an adaptive content-based image retrieval (CBIR) system for improving the performance of image retrieval in mammographic databases. In this work, the authors propose a new relevance feedback approach based on incremental learning with support vector machine (SVM) regression. Also, the authors present a new local perturbation method to further improve the performance of the proposed relevance feedback system. The approaches enable efficient online learning by adapting the current trained model to changes prompted by the user's relevance feedback, avoiding the burden of retraining the CBIR system. To demonstrate the proposed image retrieval system, the authors used two mammogram data sets: A set of 76 mammograms scored based on geometrical similarity and a larger set of 200 mammograms scored by expert radiologists based on pathological findings. The experimental results show that the proposed relevance feedback strategy improves the retrieval precision for both data sets while achieving high efficiency compared to offline SVM. For the data set of 200 mammograms, the authors obtained an average precision of 0.48 and an area under the precision-recall curve of 0.79. In addition, using the same database, the authors achieved a high pathology matching rate greater than 80% between the query and the top retrieved images after relevance feedback. Using mammographic databases, the results demonstrate that the proposed approach is more accurate than the model without using relevance feedback not only in image retrieval but also in pathology matching while maintaining its effectiveness for online relevance feedback applications.

Image retrieval and image compression are both very active fields of research. Unfortunately, in the past they were pursued independently leading to image indexing methods being both efficient and effective but restricted to uncompressed images. In this paper we introduce an image retrieval technique that operates in the compressed domain of vector quantize images. Vector quantization (VQ) achieves compression by representing image blocks as indices into a codebook of prototype blocks. By realizing that, if images are coded with their own VQ codebook then much of the image information is contained in the codebook itself, we propose the comparison of the codebooks, based on a Modified Hausdorff distance, as a novel method for compressed domain image retrieval. Experiments, based on an image database comprising many colorful pictures show this technique to give excellent results, outperforming classical color indexing techniques.

In recent years, secure image archival and retrieval model in cloud computing (CC) received significant attention to ensure data confidentiality and secured data transmission among cloud server and clients cloud storage and users. Traditional secure image retrieval (IR) techniques are not applicable to adapt with a large-scale IR in cloud environment. In order to overcome these issues, this article introduces an effective secure IR scheme using Inception with ResNet v2 (SIRS-IR) and Multiple Share Creation (MSC) scheme. The proposed method uses Inception with ResNet v2 model based feature extraction process. At the same time, the MSC process is utilized for multiple share generation and then encryption of shares takes place using double chaotic logistic map (DLCM) technique. Besides, the encrypted shares and the respective feature vectors are saved in the cloud server with the corresponding image identification number. During the IR process for the applied QI, the SIRS-IR model extracts the feature vectors and performs similarity measurement to retrieve the related images from the database interms of encrypted shares. Finally, the share decryption process is carried out for the reconstruction of original images with no loss of quality. Extensive experimentations were performed to verify the retrieval performance and image quality of the reconstructed images using Corel10K dataset. The obtained results stated that the presented SIRS-IR model is found to be superior to other methods in a considerable way.

DNA is emerging as a promising medium for storing huge volumes of data in a confined space that remains intact for thousands of years. Although this technique is very efficient, especially for multimedia data like images, there is a lack of efficient searching and retrieval technique. This paper addresses this issue and proposes a novel Content Based Image Retrieval (CBIR) technique to retrieve similar images from the generated DNA-based image feature vectors. The features are obtained by a novel encoding scheme that uses the three Most-Significant Bits from the images and converts them into a string of nucleotides that follow run length and GC constraints to form DNA planes stored in a DNA medium. The nucleotides in these planes are interpreted through three consecutive sequences forming codons. The codon-based features are then utilized to perform instance-based image retrieval. The DNA planes are further adapted and implemented on diverse deep learning architectures, including ResNet-50, VGG-16, VGG-19, and Inception V3, to facilitate classification-based image retrieval tasks. The system's performance has been assessed using a range of datasets, encompassing coral, medical, and multi-label images. Experimental results demonstrate that the proposed approach achieves notable improvements when compared to existing state-of-the-art methods.

To reduce the storage and transmission requirements in image applications, images are compressed while maintaining acceptable visual quality. In this paper, image retrieval using Vector Quantization (VQ) is explored. The VQ technique is tested on uncompressed images (UCID) as well as compressed images (JPG). In order to reduce the retrieval time, the images are converted into thumbnails and the retrieval performance in both the domains is observed. The testing is performed with various thumbnail sizes. Performance is evaluated by using Precision and Recall. It is observed that the retrieval performance using VQ for compressed images outperforms that with uncompressed images.