Multimedia Database Research Articles

An efficient retrieval using edge GLCM and association rule mining guided IPSO based artificial neural network

Content Based Image Retrieval (CBIR) is a challenging research area due to increase in multimedia database and other image libraries day by day. With an intent to provide an efficient search and retrieval, we propose an enhanced Content Based Medical Image Retrieval (CBMIR) system to support the medical practitioners in their diagnosis task. For which, we introduce boosted feature extraction and retrieval phase for medical images using Edge GLCM (EGLCM) and Association Rule Mining (ARM) integrated with Artificial Neural Network (ANN). Improved Particle Swarm Optimization (IPSO) is deployed to optimize the weights of ANN. The system is put forth with four important phases; 1. Pre-Processing, 2. Feature Extraction using Edge Histogram Descriptor (EHD), Local Gabor XOR Pattern (LGXP) and EGLCM, 3. Association Rule Mining using Apriori and 4. Optimized Retrieval using IPSO based ANN and Euclidean distance. In ANN, 7,000 images are trained and 1,100 images are tested. On Comparison with the existing systems, our method has shown best results with improved accuracy of 95 % in addition to reduced computational complexity by pre-processing and dimensionality reduction through minimal feature vector.

Multimodal Discriminative Binary Embedding for Large-Scale Cross-Modal Retrieval.

Multimodal hashing, which conducts effective and efficient nearest neighbor search across heterogeneous data on large-scale multimedia databases, has been attracting increasing interest, given the explosive growth of multimedia content on the Internet. Recent multimodal hashing research mainly aims at learning the compact binary codes to preserve semantic information given by labels. The overwhelming majority of these methods are similarity preserving approaches which approximate pairwise similarity matrix with Hamming distances between the to-be-learnt binary hash codes. However, these methods ignore the discriminative property in hash learning process, which results in hash codes from different classes undistinguished, and therefore reduces the accuracy and robustness for the nearest neighbor search. To this end, we present a novel multimodal hashing method, named multimodal discriminative binary embedding (MDBE), which focuses on learning discriminative hash codes. First, the proposed method formulates the hash function learning in terms of classification, where the binary codes generated by the learned hash functions are expected to be discriminative. And then, it exploits the label information to discover the shared structures inside heterogeneous data. Finally, the learned structures are preserved for hash codes to produce similar binary codes in the same class. Hence, the proposed MDBE can preserve both discriminability and similarity for hash codes, and will enhance retrieval accuracy. Thorough experiments on benchmark data sets demonstrate that the proposed method achieves excellent accuracy and competitive computational efficiency compared with the state-of-the-art methods for large-scale cross-modal retrieval task.

Active rule base development for dynamic vertical partitioning of multimedia databases

Currently, vertical partitioning has been used in multimedia databases in order to take advantage of its potential benefits in query optimization. Nevertheless, most vertical partitioning algorithms are static; this means that they optimize a vertical partitioning scheme (VPS) according to a workload, but if this workload suffers changes, the VPS may be degraded, which would result in long query response time. This paper presents a set of active rules to perform dynamic vertical partitioning in multimedia databases. First of all, these rules collect all the information that a vertical partitioning algorithm needs as input. Then, they evaluate this information in order to know if the database has experienced enough changes to trigger a performance evaluator. In this case, if the performance of the database falls below a threshold previously calculated by the rules, the vertical partitioning algorithm is triggered, which gets a new VPS. Finally, the rules materialize the new VPS. Our active rule base is implemented in the system DYMOND, which is an active rule-based system for dynamic vertical partitioning of multimedia databases. DYMOND's architecture and workflow are presented in this paper. Moreover, a case study is used to clarify and evaluate the functionality of our active rule base. Additionally, authors of this paper performed a qualitative evaluation with the aim of comparing and evaluating DYMOND's functionality. The results showed that DYMOND improved query performance in multimedia databases.

Efficient indexing of binary LSH for high dimensional nearest neighbor

Approximate Nearest Neighbor search (ANN) is one of the most frequently used and yet expensive operations in the high-dimensional database, especially the multimedia database involving massive high-dimensional feature vectors. Recently, Locality-Sensitive Hashing and its variants have been generally acknowledged as the promising solutions to ANN search due to its excellent performance and ease of implementation. However, the existing LSH methods for external memory usually create a linear order relationship on the set of compound hash keys to rearrange the data set in the external storage, which incurs many false negatives.In this paper, we propose a surprising simple method to solve the ANN problem with high accuracy results and requiring only a limited number of random I/O. For the distance-preserving properties of LSH functions, the idea of collision counting is used to guarantee the accuracy of the returned neighbours. We convert the dynamic collision counting problem into the nearest neighbor search (NN) in Hamming space. To support the NN search in the hamming space, we establish the multi-index hashing structure to rearrange the binary hash keys and their corresponding objectives. Accessing the candidates consumes a limited number of random I/O and shorter response time. Experimental results show that our method can achieve higher accuracy of ANN results compared with the state-of-the-art methods including LSB, C2LSH and SK-LSH.

New dynamic metric indices for secondary memory

Metric indices support efficient similarity searches in metric spaces. This problem is central to many applications, including multimedia databases and repositories handling complex objects. Most metric indices are designed for main memory, and also most of them are static, that is, do not support insertions and deletions of objects. In this paper we introduce new metric indices for secondary memory that support updates, that is, they are dynamic. First, we show how the dynamic and memory-based Dynamic Spatial Approximation Tree (DSAT) can be extended to operate on secondary memory. Second, we design a dynamic and secondary-memory-based version of the static List of Clusters (LC), which performs well on high-dimensional spaces. The new structure is called Dynamic LC (DLC). Finally, we combine the DLC with the in-memory version of DSAT to create a third structure, Dynamic Set of Clusters (DSC), which improves upon the other two in various cases. We compare the new structures with the state of the art, showing that they are competitive and outstand in several scenarios, especially on spaces of medium and high dimensionality.

Multimodal Data Mining in a Multimedia Database Based on Structured Max Margin Learning

Mining knowledge from a multimedia database has received increasing attentions recently since huge repositories are made available by the development of the Internet. In this article, we exploit the relations among different modalities in a multimedia database and present a framework for general multimodal data mining problem where image annotation and image retrieval are considered as the special cases. Specifically, the multimodal data mining problem can be formulated as a structured prediction problem where we learn the mapping from an input to the structured and interdependent output variables. In addition, in order to reduce the demanding computation, we propose a new max margin structure learning approach called Enhanced Max Margin Learning (EMML) framework, which is much more efficient with a much faster convergence rate than the existing max margin learning methods, as verified through empirical evaluations. Furthermore, we apply EMML framework to develop an effective and efficient solution to the multimodal data mining problem that is highly scalable in the sense that the query response time is independent of the database scale. The EMML framework allows an efficient multimodal data mining query in a very large scale multimedia database, and excels many existing multimodal data mining methods in the literature that do not scale up at all. The performance comparison with a state-of-the-art multimodal data mining method is reported for the real-world image databases.

Effective 3-D surface modeling for geographic information systems

Abstract. In this work, we propose a dynamic, flexible and interactive urban digital terrain platform with spatial data and query processing capabilities of geographic information systems, multimedia database functionality and graphical modeling infrastructure. A new data element, called Geo-Node, which stores image, spatial data and 3-D CAD objects is developed using an efficient data structure. The system effectively handles data transfer of Geo-Nodes between main memory and secondary storage with an optimized directional replacement policy (DRP) based buffer management scheme. Polyhedron structures are used in digital surface modeling and smoothing process is performed by interpolation. The experimental results show that our framework achieves high performance and works effectively with urban scenes independent from the amount of spatial data and image size. The proposed platform may contribute to the development of various applications such as Web GIS systems based on 3-D graphics standards (e.g., X3-D and VRML) and services which integrate multi-dimensional spatial information and satellite/aerial imagery.

Similarity (range and kNN) queries processing on an Intel Xeon Phi coprocessor

Nowadays, the evolution of information technologies requires fast similarity search tools for analyzing new data types as audio, video, or images. The usual search by keys or records is not possible and to search on these databases is a compute-intensive problem. Regarding this, in the latest years, compute-intensive coprocessors (mainly NVIDIA GPUs) have been studied as a tool for accelerating sequential processing algorithms. In this work, we implement kNN and range queries on the recently launched Intel Xeon Phi coprocessor. We developed exhaustive and also indexing algorithms using the LC index. This index has been widely studied in sequential computing to accelerate similarity search on multimedia databases. We implement and compare different exhaustive and indexing versions showing some key factors in Xeon Phi to deal with this type of search. For indexing algorithms, we used a strategy based on cluster distribution among cores LC MIC Dist-C obtaining up to 168$$\times $$× over the sequential exhaustive algorithm. Our algorithms using exhaustive strategies in Xeon Phi for range queries achieve up to 22$$\times $$× speed-up over the sequential counterpart compared to the 12$$\times $$× of a 20-core machine, and a similar advantage is achieved for kNN queries. Comparing with GPUs, we obtain higher performance on our indexing algorithms on Intel Xeon Phi. However, GPU works faster with memory-aligned access exhaustive algorithms. Our exhaustive approaches on Xeon Phi can be used on a wide class of databases, for example, non-metric spaces. Finally, we extend our algorithms to be used with large databases that do not fit in the coprocessor memory, showing a good scalability with the number of elements.

Exact and approximate flexible aggregate similarity search

Aggregate similarity search, also known as aggregate nearest-neighbor (Ann) query, finds many useful applications in spatial and multimedia databases. Given a group Q of M query objects, it retrieves from a database the objects most similar to Q, where the similarity is an aggregation (e.g., $${{\mathrm{sum}}}$$sum, $$\max $$max) of the distances between each retrieved object p and all the objects in Q. In this paper, we propose an added flexibility to the query definition, where the similarity is an aggregation over the distances between p and any subset of $$\phi M$$?M objects in Q for some support$$0< \phi \le 1$$0<?≤1. We call this new definition flexible aggregate similarity search and accordingly refer to a query as a flexible aggregate nearest-neighbor ( Fann ) query. We present algorithms for answering Fann queries exactly and approximately. Our approximation algorithms are especially appealing, which are simple, highly efficient, and work well in both low and high dimensions. They also return near-optimal answers with guaranteed constant-factor approximations in any dimensions. Extensive experiments on large real and synthetic datasets from 2 to 74 dimensions have demonstrated their superior efficiency and high quality.

Implementing watermarking as multimedia database security using hash partitioning with combination ga and hbo

Implementing watermarking as multimedia database security using hash partitioning with combination ga and hbo

Effective and Efficient Algorithms for Flexible Aggregate Similarity Search in High Dimensional Spaces

Numerous applications in different fields, such as spatial databases, multimedia databases, data mining, and recommender systems, may benefit from efficient and effective aggregate similarity search, also known as aggregate nearest neighbor (AggNN) search. Given a group of query objects $Q$ , the goal of AggNN is to retrieve the $k$ most similar objects from the database, where the underlying similarity measure is defined as an aggregation (usually sum or max ) of the distances between the retrieved objects and every query object in $Q$ . Recently, the problem was generalized so as to retrieve the $k$ objects which are most similar to a fixed proportion of the elements of $Q$ . This variant of aggregate similarity search is referred to as “flexible AggNN”, or FANN. In this work, we propose two approximation algorithms, one for the sum variant of FANN, and the other for the max variant. Extensive experiments are provided showing that, relative to state-of-the-art approaches (both exact and approximate), our algorithms produce query results with good accuracy, while at the same time being very efficient.

Cache locality is not enough

Nearest Neighbor (NN) search in high dimension is an important feature in many applications (e.g., image retrieval, multimedia databases). Product Quantization (PQ) is a widely used solution which offers high performance, i.e., low response time while preserving a high accuracy. PQ represents high-dimensional vectors (e.g., image descriptors) by compact codes. Hence, very large databases can be stored in memory, allowing NN queries without resorting to slow I/O operations. PQ computes distances to neighbors using cache-resident lookup tables, thus its performance remains limited by (i) the many cache accesses that the algorithm requires, and (ii) its inability to leverage SIMD instructions available on modern CPUs. In this paper, we advocate that cache locality is not sufficient for efficiency. To address these limitations, we design a novel algorithm, PQ Fast Scan, that transforms the cache-resident lookup tables into small tables, sized to fit SIMD registers. This transformation allows (i) in-register lookups in place of cache accesses and (ii) an efficient SIMD implementation. PQ Fast Scan has the exact same accuracy as PQ, while having 4 to 6 times lower response time (e.g., for 25 million vectors, scan time is reduced from 74ms to 13ms).

Construction strategies of teaching database for blood disease testing technology

According to the practicality, openness, sustainable development and resource diversity principle, we have constructed the teaching database for blood disease testing technology with text library, courseware database and multimedia database as the main content. We have also put forward the strategies that can be used in the construction of the teaching database. First is unified coordination and scientific management. That is the principal appoints Office of Academic Affairs to be responsible for the organization and coordination, planning and implementation, the development team teaching database specific division of labor and the task to the people. The second is based on the platform of digital campus to construct the teaching database, which can provide the platform for the upload network, running and management of teaching resources. The third is cooperation of colleges and universities to build the teaching database for blood disease testing technology, which is industry professionals demand oriented for the service of students' future jobs. Key words: Blood disease testing technology; Teaching database; Construction principle; Construction strategy

A scalable summary generation method based on cross-modal consensus clustering and OLAP cube modeling

Video summarization has been a core problem to manage the growing amount of content in multimedia databases. An efficient video summary should display an overview of the video content and most exis...

Teaching Resources-based Multimedia Database and Its Application

Teaching Resources-based Multimedia Database and Its Application

Automatic Mapping Magnetic Resonance Images into Multimedia Database Using SIFT

This paper focuses on the representation of magnetic resonances of different parts of the human body, such as knees, spinal column, arms, elbows, etc., using ontologies. First, it maps the resonance images in a multimedia database. Then, automatically, using the SIFT pattern recognition algorithm, descriptors of the images stored in the database are extracted in order to recover useful data for the user; it uses the ontologies as an artificial intelligence tool and, in consequence, reduces generation of useless data. Why do we think this is an interesting task? Because, if the user requires information about any topics or (s)he has some illness or needs to undergo magnetic resonance, this tool will show him/her images and text to convey a better understanding, helping to obtain useful conclusions. Artificial intelligence techniques are used, such as machine learning, knowledge representation, and pattern recognition. The ontological relations introduced here are based on the common representation of language, using definition dictionaries, Roget's thesaurus, synonym dictionaries, and other resources The system generates an output in the OM ontological language [1]. This language represents a structure where our system adds the data scanned by the SIFT algorithm. The tests have been made in Spanish; however, thanks to the portability of our system, it is possible to extend the method to any language.

Media Education in Russia: Past and Present

Within the framework of conferences the reports directly concerning questions of media education, problems of the organization of multimedia databases, electronic libraries, and mediateques in libraries for children and youth were heard. Important objective for Russian media educators is to open (get it registered by the Russian Ministry of Education and Sciences) a new university major speciality (major) "Media Education" within the framework of which it will be possible to prepare professional media educators for universities and schools. Within the context of increasing interest to media education worldwide, the UNESCO program's support, recent developments such as the introduction of a pre-service teacher training, and the systematic publication of a journal, media education has good prospects in Russia.

Efficient extraction of clustering-based feature signatures using GPU architectures

Similarity search and content-based retrieval have become widely used in multimedia database systems that often manage huge data collections. Unfortunately, many effective content-based similarity ...

The Folklore Archive of the Russian State University for the Humanities in MoscowFolklorni arhiv Ruske državne univerze za humanistične vede v Moskvi

The subject of the article is the field research of the folk tradition of the northern Russian area (the region of Kargopol, Archangelskaya oblast), led by the folklore and ethnolinguistic expedition of the Russian State University for the Humanities (RSUH) since 1993 until now, and the expedition data archive. The expeditions are organized 1–3 times a year, the participants include professors, postgraduate and undergraduate students of the RSUH. Several groups take part in the expedition, each consisting of 10–15 people who investigate one village during 2–4 weeks. The expedition is aimed at collecting ethnographic, ethnolinguistic and folkloric data, which could give a researcher the opportunity to get as detailed as possible a description of a modern folk culture state in the region. The main method of the field research used during the expedition work is interviewing a considerable part of the village inhabitants – about 20-60% – using 27 questionnaires that concern different spheres of the traditional culture, the audio records are later deciphered, and deciphered texts make the card index. The computer version of the archive has been created in the Laboratory of Folklore since 1997. The multimedia database “Traditional culture of Northern Russia (Kargopol region)” contains written texts, audio records of some fragments of interviews and graphic data (photos, scanned manuscripts, containing incantations, prayers, songs, etc). The database includes now about 20% of all archive materials. The card index contains about 120000 cards, while the database – about 23000 cards.

EVOLUTION OF CROWD-SOURCING FROM COMPUTER TO MOBILE SYSTEMS

As crowd-sourcing is becoming popular for problem solving and completing a task, it is now very important to use this concept in an advance manner. It can also be used as a distributed and vast source of information. This concept is now evolving in world of mobile systems. This will be a little different from that of computer systems. In this paper, we have discussed some new technologies and challenges before us to implement these advancements in crowd-sourcing. We are going to talk about cheat-detection techniques, handling multimedia databases and how to trade off between cost and accuracy by considering the redundant data as well.