Content-based Search Research Articles

Confidentiality-Preserving Image Search: A Comparative Study Between Homomorphic Encryption and Distance-Preserving Randomization

Recent years have seen increasing popularity of storing and managing personal multimedia data using online services. Preserving confidentiality of online personal data while offering efficient functionalities thus becomes an important and pressing research issue. In this paper, we study the problem of content-based search of image data archived online while preserving content confidentiality. The problem has different settings from those typically considered in the secure computation literature, as it deals with data in rank-ordered search, and has a different security-efficiency requirement. Secure computation techniques, such as homomorphic encryption, can potentially be used in this application, at a cost of high computational and communication complexity. Alternatively, efficient techniques based on randomizing visual feature and search indexes have been proposed recently to enable similarity comparison between encrypted images. This paper focuses on comparing these two major paradigms of techniques, namely, homomorphic encryption-based techniques and feature/index randomization-based techniques, for confidentiality-preserving image search. We develop novel and systematic metrics to quantitatively evaluate security strength in this unique type of data and applications. We compare these two paradigms of techniques in terms of their search performance, security strength, and computational efficiency. The insights obtained through this paper and comparison will help design practical algorithms appropriate for privacy-aware cloud multimedia systems.

Content-based Search for Image Retrieval

In this paper, a content-based image retrieval approach is presented for effective searching. The proposed approach uses two or more types of query for accessing images, textual annotation associated with images and visual appearance, such as colour, texture and positional features of objects in sample images. One can first place a keyword-based query, and then the desired images are retrieved by visual content-based query. The proposed retrieval approach shows clear improvements over competing approaches in terms of retrieval accuracy and visual inspection using Corel gallery and WWW images.

A Look into Content-Based Image Search Engine System

Image search engine technology is a combination of technologies of database, information retrieval, image processing and computer vision, texture recognition as well as multimedia database and network. This paper aims to present the flow chart of content-based image search engine system and explores the current image search technologies and its future development.

The EHR-ARCHE project: Satisfying clinical information needs in a Shared Electronic Health Record System based on IHE XDS and Archetypes

PurposeWhile contributing to an improved continuity of care, Shared Electronic Health Record (EHR) systems may also lead to information overload of healthcare providers. Document-oriented architectures, such as the commonly employed IHE XDS profile, which only support information retrieval at the level of documents, are particularly susceptible for this problem. The objective of the EHR-ARCHE project was to develop a methodology and a prototype to efficiently satisfy healthcare providers’ information needs when accessing a patient's Shared EHR during a treatment situation. We especially aimed to investigate whether this objective can be reached by integrating EHR Archetypes into an IHE XDS environment. MethodsUsing methodical triangulation, we first analysed the information needs of healthcare providers, focusing on the treatment of diabetes patients as an exemplary application domain. We then designed ISO/EN 13606 Archetypes covering the identified information needs. To support a content-based search for fine-grained information items within EHR documents, we extended the IHE XDS environment with two additional actors. Finally, we conducted a formative and summative evaluation of our approach within a controlled study. ResultsWe identified 446 frequently needed diabetes-specific information items, representing typical information needs of healthcare providers. We then created 128 Archetypes and 120 EHR documents for two fictive patients. All seven diabetes experts, who evaluated our approach, preferred the content-based search to a conventional XDS search. Success rates of finding relevant information was higher for the content-based search (100% versus 80%) and the latter was also more time-efficient (8–14min versus 20min or more). ConclusionsOur results show that for an efficient satisfaction of health care providers’ information needs, a content-based search that rests upon the integration of Archetypes into an IHE XDS-based Shared EHR system is superior to a conventional metadata-based XDS search.

Visual word expansion and BSIFT verification for large-scale image search

Recently, great advance has been made in large-scale content-based image search. Most state-of-the-art approaches are based on the bag-of-visual-words model with local features, such as SIFT, for image representation. Visual matching between images is obtained by vector quantization of local features. Feature quantization is either performed with hierarchical k-NN which introduces severe quantization loss, or with ANN (approximate nearest neighbors) search such as k-d tree, which is computationally inefficient. Besides, feature matching by quantization ignores the vector distance between features, which may cause many false-positive matches. In this paper, we propose constructing a supporting visual word table for all visual words by visual word expansion. Given the initial quantization result, multiple approximate nearest visual words are identified by checking supporting visual word table, which benefits the retrieval recall. Moreover, we present a matching verification scheme based on binary SIFT (BSIFT) signature. The L2 distance between original SIFT descriptors is demonstrated to be well kept with the metric of Hamming distance between the corresponding binary SIFT signatures. With the BSIFT verification, false-positive matches can be effectively and efficiently identified and removed, which greatly improves the precision of large-scale image search. We evaluate the proposed approach on two public datasets for large-scale image search. The experimental results demonstrate the effectiveness and efficiency of our scheme.

Performance Comparison of Wavelet Transform with respect to Their Transforms of CBVR

Content-based search and retrieval of video data has become a challenging and important issue. Different methods[1,2,3] have been proposed to develop video retrievals systems to achieve better performance in terms of accuracy. The proposed technique uses wavelettransforms that areCosine Wavelet Transform, Walsh, Kekre, Slant and Hartley Wavelet Transform. The benefit of energy compaction of fewer transformedcoefficients is taken to reduce the feature vector size by taking fractional coefficients[5]of transformed frames of video. Smaller feature vector size results in less time for comparison of feature vectors resulting in faster retrieval of images. The feature vectors are extracted and coefficients sets are considered as feature vectors (100%, 6.25%, 3.125%, 1.5625%, 0.7813%, 0.39%, 0.195%, 0.097%, 0.048%, 0.024%, 0.012%, 0.006% and 0.003% of complete transformed coefficients).The database consists of 500 videos spread across 10 categories.Performance using wavelets is compared with respect to their transforms. In all considered wavelet transform, experimental result show the performance improvement in fractional coefficient of wavelet transform in content based video retrieval as compared to 100% coefficient considered.

Using Contextual Information to Assist Desktop Search

As the volume of personal data grows, it becomes increasingly difficult to rapidly find what we need on our desktop computers. Traditional content-based desktop search tools provide a potential solution by indexing file content, while lacking of contextual information about the user's activity. Context-enhanced search engines seem promising: they re-rank content-based results with the user's context. However, most of them are incomplete answers: they neither consider the factor of network application (especially, Web browser), nor take full advantage of the characteristics of user behavior and file metadata on desktop computers. To address those issues, we propose a concept of ”document-window switch tree” to represent the dynamic relationship from the user's context. We further collect relevant information to re-rank traditional content search results, such as reading time, frequency of window switch, file content and metadata. Search results on ranking accuracy generated by our method significantly outperform the context-enhanced search tool based on temporal locality.

Mejorando las búsquedas en <i>Europeana</i>, el proyecto <i>Assets</i>

Since its launch several years ago, Europeana has continued to improve the quantity and quality of its content, attempting to offer new services and differentiate itself from the more popular search engines. The Assets project has contributed to this goal by developing innovative content-based search technologies and services, improving existing functionality, and designing more user-friendly interfaces. This paper describes in detail some of the new services that were implemented in a replica of the Europeana website. A user evaluation process was conducted in several European countries, with a total of 51 usability tests performed. The results reveal the benefits and drawbacks of the portal. The identified problems were systematically prioritized. In the final phase of the project, a redesign resolved 74% of all usability incidents. The result is a number of features that enhance the original Europeana portal, giving the user more and better search tools.

Perceptual audio hashing algorithm based on Zernike moment and maximum-likelihood watermark detection

A new perceptual audio hashing algorithm based on maximum-likelihood watermarking detection is proposed in this paper. The idea is justified by the fact that the maximum-likelihood watermark detector responds similarly to perceptually close audio using a non-embedded watermark (i.e. virtual watermark). The feature vector, which is composed of the total amplitude of low-order Zernike moments of each audio frame, is modeled by the Gaussian or Rayleigh distribution. Then, the maximum-likelihood watermark detection is performed on the feature vector with the virtual watermarks generated by pseudo-random number generator to construct the hash vector. Extensive experiments over three large audio databases of different type (speech, instrumental music, and sung voice) demonstrate the efficiency of the proposed scheme in terms of discrimination, perceptual robustness and identification rate. It is also verified that the proposed scheme outperforms state-of-the-art techniques in perceptual robustness and can be applied in content-based search, successfully.

Dominant Color and Texture Approached for Content Based Video Images Retrieval

Content-based retrieval allows finding information by searching its content rather than its attributes. Content-based search and retrieval of video data becomes a challenging and important problem. Every year video content is growing in volume and there are different techniques available to capture, compress, display, store and transmit video while editing and manipulating video based on their content is still a non-trivial activity. Recent advances in multimedia technologies allow the capture and storage of video data with relatively inexpensive computers. However, without appropriate search techniques all these data are hardly usable. Today research is focused on video retrieval.Moreover, content-based video retrieval system requires first of all segment the video stream into separate shots. Video Shot Afterwards features are extracted for video shots representation. And finally, choose a similarity/distance metric and an algorithm that is efficient enough to retrieve query - related videos results. There are two main issues in this process; the first is how to determine the best way for video segmentation and key frame selection. The second is the features used for video representation. Various features can be extracted for this sake including either low or high level features. A key issue is how to bridge the gap between low and high level features. In this paper we presented approach for content based video retrieval based on Dominant color and texture of a video image. We also talk about video Representation, feature extraction from like texture, dominant color and color histogram from video frame. Keywords- Video retrieval, dominant color, Gray level co occurrence matrix. Feature extraction, Key frame extraction, Video representation, and Video segmentation. Image Retrieval, color Histogram

Research and Design for Image Search Engine

Content-based image search is an urgent problem as the next generation of search engines. This paper firstly analyzes and discusses its main features and key techniques, and then presents a designs method of image search engine based on image content. Next, we give a detailed elaboration about the main function modules and introduce the testing process. By using data filtering technology and ELFHash algorithms, the search time-consume is less than 1 second. The testing results finally prove the overall performance of our image search engine is excellent and achieves the desired design requirements.

Satellite Image Retrieval using Modified Block Truncation Coding and Kekre Transform Patterns

Satellite images and Content Based Image Retrieval are interesting topics of research since years. Specifically, CBIR is on developing technologies for bridging the semantic gap that currently prevents wide-deployment of image content-based search engines. Image search engines currently in use are mostly rely on human generated data, such as text. Annotation of an image is totally depend on the person’s perception who is going to store it into database. It is time-consuming as well as error prone. Therefore search engine using text input results in various non-relevant images. To overcome drawbacks of text based image retrieval, Content based image retrieval is introduced where retrieval of images is totally depend on the features of images. Mostly, content-based methods are based on low-level descriptions, while high-level or semantic descriptions are beyond current capabilities. In this paper, we will try to implement the technique to fill this gap. This technique can eventually be extended to allow for content-based similarity type of search, to find out different query blocks from satellite image. We will fire an object as a query and most similar blocks will get retrieved. This will help out to trace particular object from large satellite image.

Content-based search overviews and exploratory browsing of movies with MovieClouds

Videos, and especially movies, often engage viewers perceptually, cognitively and emotionally, by combining diverse symbol systems, such as images, texts, music and narration to tell stories. As one of the biggest sources of entertainment, in individual and social contexts, they are increasingly accessible as enormous collections over the internet, in social media and iTV. These richer environments demand for new and more powerful ways to search, browse and view videos that may benefit from video content-based analysis and classification techniques. In this paper, we describe and evaluate MovieClouds, its core and extended features of content processing, interactive search, overview and browsing designed to access, explore and visualise movies, from overview clouds, at the movies space down to the movies, based on the information conveyed in their different content tracks. It adopts a tag cloud unifying-paradigm, to extend to movies the power, flexibility, engagement and fun usually associated with clouds.

Shot Detection Using Genetic Edge Histogram and Object Based Video Retrieval Using Multiple Features

As the usage of multimedia data increasing rapidly, how to get the video data we need efficiently become so important. Recent advances in multimedia technologies allow the capture and storage of video data with relatively inexpensive computers. Problem Statement: However, without appropriate search techniques all these data are hardly usable. Users want to query the content instead of the raw video data. Today research is focused on video retrieval. Content-based search and retrieval of video data becomes a challenging and important problem. To retrieve the content of the video the user need automatic classification and categorization of the visual content. Approach: In this study a novel algorithm is proposed for shot detection using Genetic Edge Histogram and 2-D discreate cosine transform as a feature and multiple features like color, motion, shape and SIFT are used to retrieve the similar shots. Results and Conclusion: The combination of proposed features yields good results interms of precision and recall.

Finding and reusing learning materials with multimedia similarity search and social networks

The authors describe how content-based multimedia search technologies can be used to help learners find new materials and learning pathways by identifying semantic relationships between educational resources in a social learning network. This helps users – both learners and educators – to explore and find material to support their learning aims. Exciting new technologies are emerging for designing and deploying multi-modal, interactive educational tools such as video, 3D models, games, remote sensors or collaborative community-created resources (e.g. wikis). With the increasing availability of open educational resources in online, semantically marked-up and socially connected spaces, users are building their own learning pathways by exploring and remixing content. In this article the authors look at how the integration of multimedia search into the SocialLearn platform for shared learning spaces can contribute to this vision.

Research and Improvement on Content-Based Web Search Engine

There is a lot of information contained in the World Wide Web. It has become a research focus to obtain the required related resources quickly and accurately from the web through the content-based search engines. Most current tools of full text web search engine, such as Lucene which is a widely used open source retrieval library in information retrieval field, are purely keyword based. This may not sufficient for users to retrieve in the web. In this paper, we employ a method to overcome the limitations of current full text search engines in represent of Lucene. We propose a Query Expansion and Information Retrieval approach which can help users to acquire more accurate contents from the web. The Query Expansion component finds expanded candidate words of the query word through WordNet which contains synonyms in several different senses; In the Information Retrieval component, the query word and its candidate words are used together as the input of the search module to get the result items. Furthermore, we can put the result items into different classes based on the expansion. Some experiments and the results are described in the late part of this paper.

IPixel: A visual content-based and semantic search engine for retrieving digitized mammograms by using collective intelligence

Nowadays, traditional search engines such as Google, Yahoo and Bing facilitate the retrieval of information in the format of images, but the results are not always useful for the users. This is mainly due to two problems: (1) the semantic keywords are not taken into consideration and (2) it is not always possible to establish a query using the image features. This issue has been covered in different domains in order to develop content-based image retrieval (CBIR) systems. The expert community has focussed their attention on the healthcare domain, where a lot of visual information for medical analysis is available. This paper provides a solution called iPixel Visual Search Engine, which involves semantics and content issues in order to search for digitized mammograms. iPixel offers the possibility of retrieving mammogram features using collective intelligence and implementing a CBIR algorithm. Our proposal compares not only features with similar semantic meaning, but also visual features. In this sense, the comparisons are made in different ways: by the number of regions per image, by maximum and minimum size of regions per image and by average intensity level of each region. iPixel Visual Search Engine supports the medical community in differential diagnoses related to the diseases of the breast. The iPixel Visual Search Engine has been validated by experts in the healthcare domain, such as radiologists, in addition to experts in digital image analysis.

JPSearch: New international standard providing interoperable framework for image search and sharing

For the last 15 years, there have been various research and great advances in the field of content-based search and retrieval of images. In this paper, we present an overview of the JPSearch Standard, which is a recently published set of international standards providing interoperable framework for image search and sharing. The JPSearch standard is composed of six parts, which are Part 1: System framework and components, Part 2: Registration, identification and management of schema and ontology, Part 3: Query format, Part 4: File format for metadata embedded in image data (JPEG and JPEG 2000), Part 5: Data interchange format between image repositories, Part 6: Reference software. The Part 1 provides motivation and overview of the JPSearch framework. The Part 2 supports interoperability among various metadata specifications and social tagging using the specification of core metadata and translation rule description language. The Part 3 provides powerful image query language for interoperability among multiple image databases. The Part 4 specifies a file format based on JPEG and JPEG 2000 file format in which arbitrary number of metadata description of a image can be embedded. The Part 5 supports interoperability in interchanging images with metadata between image repositories. Finally, the Part 6 provides reference and utility softwares based on which the JPSearch compliant systems can be easily developed. In this paper, summaries and usage examples for each part of the JPSearch standard are presented.

Discriminative codebook learning for Web image search

Given the explosive growth of the Web images, image search plays an increasingly important role in our daily lives. The visual representation of image is the fundamental factor to the quality of content-based image search. Recently, bag-of-visual word model has been widely used for image representation and has demonstrated promising performance in many applications. In the bag-of-visual-word model, the codebook/visual vocabulary plays a crucial role. The conventional codebook, generated via unsupervised clustering approaches, does not embed the labeling information of images and therefore has less discriminative ability. Although some research has been conducted to construct codebooks with the labeling information considered, very few attempts have been made to exploit manifold geometry of the local feature space to improve codebook discriminative ability. In this paper, we propose a novel discriminative codebook learning method by introducing the subspace learning in codebook construction and leveraging its power to find a contextual local descriptor subspace to capture the discriminative information. The discriminative codebook construction and contextual subspace learning are formulated as an optimization problem and can be learned simultaneously. The effectiveness of the proposed method is evaluated through visual reranking experiments conducted on two real Web image search datasets.

멀티미디어 콘텐츠의 강화된 의미 검색을 위한 온톨로지 설계

최근 정보 환경에서는 다양한 멀티미디어 콘텐츠가 주목받고 있다. 그러나 이러한 콘텐츠는 표현 형식과 내용이 방대하여 사용자가 원하는 콘텐츠를 쉽게 검색하고 활용하기가 어려웠다. 이를 해결하기 위해, 그동안 많은 연구들이 콘텐츠에 메타데이터를 추가하는 표준안을 제시하였고, 이로 인해 콘텐츠에 대한 의미 검색이 가능해졌다. 그러나 여러 가지 메타데이터 표준과 이를 따라 명세된 콘텐츠가 증가하면서 콘텐츠 간 상호운용성이 결여되기 시작했고, 사용자는 다시 검색의 어려움에 직면하게 되었다. 이러한 문제점을 개선하기 위해, 본 연구에서는 메타데이터 표준간의 상호운용성을 지원하고, 이를 구성하는 엘리먼트와 엘리먼트 사이에 표현된 의미 관계를 확장하여 시맨틱 검색이 강화된 TOFIC(The Ontology For Imagery Contents)을 제안한다. TOFIC에서는, MPEG-7과 TV-Anytime 표준 간 의미 관계를 분류하고 확장된 시맨틱을 정의하여 현재 정보 환경에 존재하는 멀티미디어 콘텐츠에 대한 일관된 의미 검색과 강화된 내용-기반 검색을 지원한다. In recent Information Environment, various Multimedia contents are getting noticed. But, since these contents have various formats of representation and very wide range of information, it was difficult to retrieve contents that user wanted and to utilize it. To solve these problems, many studies presented draft standards for adding metadata to contents, and then, semantic search for contents had become available. Unfortunately, as the number of metadata standards and contents increased, the lack of interoperability between contents was begun and users are faced with difficulty of search contents again. To improve these problems, this paper supports interoperability between metadata standards and expands the semantic relationship between elements and proposes an ontology which is named TOFIC(The Ontology For Imagery Contents) for enhanced semantic search. In TOFIC, the semantic relationships between MPEG-7 and TV-Anytime are classified, and extended new semantics are defined. As a result, semantic search for multimedia contents is enhanced and it is possible to retrieve most of the multimedia contents that exist on the current information environment consistently. In addition, it supports enhanced content-based search for multimedia contents.