Kasturi Chatterjee

Affinity Hybrid Tree: An Indexing Technique for Content-Based Image Retrieval in Multimedia Databases

A novel indexing and access method, called affinity hybrid tree (AH-tree), is proposed to organize large image data sets efficiently and to support popular image access mechanisms like content-based image retrieval (CBIR) by embedding the high-level semantic image-relationship in the access mechanism as it is. AH-tree combines space-based and distance-based indexing techniques to form a hybrid structure which is efficient in terms of computational overhead and fairly accurate in producing query results close to human perception. Algorithms for similarity (range and k-nearest neighbor) queries are implemented. Results from elaborate experiments are reported which depict a low computational overhead in terms of the number of I/O and distance computations and a high relevance of query results. The proposed index structure solves the existing problems of introducing high-level image relationships in a retrieval mechanism without going through the pain of translating the content-similarity measurement into feature-level equivalence and yet maintaining an efficient structure to organize the large sets of images

GeM-Tree: Towards a Generalized Multidimensional Index Structure Supporting Image and Video Retrieval

In this paper, we propose a tree-based multidimensional structure, GeM-Tree, which indexes both images and videos within a single general framework utilizing Earth Moverpsilas Distance. It can support different content-based image and video retrieval approaches, and can accommodate applications where the cross-similarity between images and videos need to be considered during content-based retrievals. Furthermore, it is flexible enough to index different video classification units and can maintain the hierarchical relationship between them. In addition, it uses a construct called hierarchical Markov model mediator to introduce high-level semantic relationships among images and different levels of video units. The experimental results indicate that GeM-Tree is a promising generalized index structure for multimedia data with low computational overhead, is flexible enough to support different retrieval approaches and generates query results with high relevance.

A NOVEL INDEXING AND ACCESS MECHANISM USING AFFINITY HYBRID TREE FOR CONTENT-BASED IMAGE RETRIEVAL IN MULTIMEDIA DATABASES

An efficient access and indexing framework, called Affinity Hybrid Tree (AH-Tree), is proposed which combines feature and metric spaces in a novel way. The proposed framework helps to organize large image databases and support popular multimedia retrieval mechanisms like Content-Based Image Retrieval (CBIR). It is efficient in terms of computational overhead and fairly accurate in producing query results close to human perception. AH-Tree, by being able to introduce the high level semantic image relationship as it is in its index structure, solves the problem of translating the content-similarity measurement into feature level equivalence which is both painstaking and error-prone. Algorithms for similarity (range and k-nearest neighbor) queries are implemented and extensive experiments are performed which produces encouraging results with low I/O and distance computations and high precision of query results.

Modeling Methodology for Component Reuse and System Integration for Hurricane Loss Projection Application

Hurricanes are one of the deadliest and perilous natural calamities on the face of earth having a severe impact both on the lives of the people and economy of a nation. Attempts have been made to mitigate hurricane aftermath, by utilizing research and tools that can analyze hurricanes and estimate projected losses. The need for such research methodologies and tools stimulated the development of a multi-disciplinary cutting edge public hurricane model called public hurricane risk and insured loss projection model (PHRLM). The complex and diverse nature of the application raises the need for module abstraction, seamless integration and effective reusability to create a uniform generic environment. Providing efficient interaction between these complex multi-disciplinary modules using different abstractions, formalisms, data formats and communications and making each module transparent enough to be reusable becomes a complicated task. This paper presents a UML based formal modeling methodology, enabling component reuse and integration of the application

A SEMANTIC INDEX STRUCTURE FOR MULTIMEDIA RETRIEVAL

To be effective multimedia retrieval mechanisms, index methods must provide not only efficient access but also meaningful retrieval by addressing challenges in multimedia retrieval. This article presents the AH+-tree, a height-balanced, tree-based index structure that efficiently incorporates high-level affinity information to support Content-Based Image Retrieval (CBIR) through similarity queries. The incorporation of affinity information allows the AH+-tree to address the problems of semantic gap and user perception subjectivity inherent to multimedia retrieval. Based on the Affinity-Hybrid Tree (AH-Tree), the AH+-tree utilizes affinity information in a novel way to eliminate the I/O overhead of the AH-Tree while maintaining the same functionality and quality of results. We explain the structure of the AH+-tree and implement and analyze algorithms for tree construction and similarity queries (range and nearest neighbor). Experimental results demonstrate the superior I/O efficiency of the AH+-tree over that of the AH-Tree and the M-tree without a detrimental impact on real-time costs of the retrieval process.

A three-dimensional geographic and storm surge data integration system for evacuation planning

The rise of offshore water caused by the high winds of a low pressure weather system, or storm surge, is a hurricanes greatest threat to human life. As weather forecasters struggle to enable coastal residents to make timely evacuation decisions, the need arises for more visually compelling and interactive storm surge visualization tools. This paper presents an interactive and three-dimensional storm surge visualization system. It integrates road, topographic, and building data to construct accurate three-dimensional models of major cities in the State of Florida. Storm surge data are then used to construct a three-dimensional ocean positioned over the terrain models. Ambient details such as wind, vegetation, ocean waves, and traffic are animated based on up-to-date wind and storm surge data. Videos of the storm surge visualizations are recorded and made available to coastal residents through a web-interface. The three-dimensional visualization of geographic and storm surge data provides a more visually compelling representation of the potential effects of storm surge than traditional two-dimensional models and is more capable to enable coastal residents to make potentially life-saving evacuation decisions.

A 3-D Traffic Animation System with Storm Surge Response

This paper presents a 3D traffic animation system that provides features of vehicle interaction and storm surge response to emulate realistic scenarios in a hurricane affected area. The proposed system constructs road objects based on a series of line segments. Vehicles on the roads are animated by keeping them coherent with the direction of the road segment and adjusting their speeds with respect to one another. The vehicle speed is controlled based upon a collision-circumventing policy and they automatically respond to flood water caused by a storm surge. The system can automatically plant lamp posts along the roads, which also respond to surge flooding. An implementation conducted with the 3D scenes on a couple of south Floridas surge-susceptible areas demonstrates the systems excellence in animating real traffic scenes and impacts of storm surge on coastal regions.

Hierarchical affinity hybrid tree: A multidimensional index structure to organize videos and support content-based retrievals

Multimedia data, especially videos, have gained enormous popularity in the recent years. Data management techniques for traditional text-based data are inadequate to handle multimedia data efficiently due to their atypical characteristics. Thus, to have a robust data management framework for complex multimedia data like videos, comparable in efficiency and capability to the traditional data management approaches, components like multimedia data storage, index, and query engines need to be developed with dedicated abilities to handle the characteristics of multimedia data like multidimensional representation and semantic gap. In this paper, we investigate the design of the second component, i.e., a multimedia index, and propose a novel tree-based multidimensional hierarchical index structure called Hierarchical Affinity Hybrid-Tree (HAH-Tree) which addresses the critical issues of multidimensionality and semantic gap. The index structure accommodates different levels of video relationships during Content-Based Video Retrieval (CBVR) by utilizing a probabilistic approach called the Hierarchical Markov Model Mediator (HMMM), which is also responsible for managing the high-level semantic content of the video components. In addition, a computationally efficient k-Nearest Neighbor (k-NN) algorithm is proposed, which supports CBVR for different video units with a high precision level.

AH+-Tree: An Efficient Multimedia Indexing Structure for Similarity Queries

This paper presents the AH+-tree, a balanced, tree-based index structure that efficiently supports Content-Based Image Retrieval (CBIR) through similarity queries. The proposed index structure addresses the problems of semantic gap and user subjectivity by considering the high-level semantics of multimedia data during the retrieval process. The AH+-tree provides the same functionality as the Affinity-Hybrid Tree (AH-Tree) but utilizes the high-level semantics in a novel way to eliminate the I/O overhead incurred by the AH-Tree due to the process of affinity propagation, which requires a complete traversal of the tree. The novel structure of the tree is explained, and detailed range and nearest neighbor algorithms are implemented and analyzed. Extensive discussions and experiments demonstrate the superior efficiency of the AH+-tree over the AH-Tree and the M-tree. Results show the AH+-tree significantly reduces I/O cost during similarity searches. The I/O efficiency of the AH+-tree and its ability to incorporate high-level semantics from different machine learning mechanisms make the AH+-tree a promising index access method for large multimedia databases.

A Distributed Multimedia Data Management over the Grid

In this chapter, we propose a distributed multimedia data management architecture, which can efficiently store and retrieve multimedia data across several nodes of a Grid environment. The main components of the proposed system comprises of a distributed multidimensional index structure, a distributed query manager handling content-based information retrievals and a load balancing technology. The proposed distributed query manager embeds the high-level semantic relationships among the multimedia data objects into the k-NN based similarity search, thus bridging the semantic gap and increasing the relevance of query results manifold. This research has two major usabilities. First, it models a web environment where each node of the Grid can be considered as the nodes or sources of data in the world-wide-web. This should help to investigate and understand the challenges and requirements of future search paradigms based on content of multimedia data rather than on text annotations, as used currently. Second, it provides the foundation to develop content-based information retrievals as a possible Grid service. Extensive experiments were conducted with varied data sizes and different number of distribution nodes. Encouraging results are obtained that makes this endeavor a potential architecture to manage complex multimedia data over a distributed environment.