High Retrieval Performance Research Articles

Investigating the Effects of Multiple Factors Towards More Accurate 3-D Object Retrieval

This paper proposes a novel framework for 3-D object retrieval, taking into account most of the factors that may affect the retrieval performance. Initially, a novel 3-D model alignment method is introduced, which achieves accurate rotation estimation through the combination of two intuitive criteria, plane reflection symmetry and rectilinearity. After the pose normalization stage, a low-level descriptor extraction procedure follows, using three different types of descriptors, which have been proven to be effective. Then, a novel combination procedure of the above descriptors takes place, which achieves higher retrieval performance than each descriptor does separately. The paper provides also an in-depth study of the factors that can further improve the 3-D object retrieval accuracy. These include selection of the appropriate dissimilarity metric, feature selection/dimensionality reduction on the initial low-level descriptors, as well as manifold learning for re-ranking of the search results. Experiments performed on two 3-D model benchmark datasets confirm our assumption that future research in 3-D object retrieval should focus more on the efficient combination of low-level descriptors as well as on the selection of the best features and matching metrics, than on the investigation of the optimal 3-D object descriptor.

Research and Design on Intranet Security File System

To solve the data security issue in intranet massive storage system, a Multi-Protocol Secure File System ( for short MPSFS) is designed. Firstly, the MPSFS supports the access of users with different protocols, and provides the unified access interface, so can achieve high performance in data storage and retrieval; secondly, with the help of other technologies such as identity authentication, access control and data encryption, the MPSFS can effectively ensure the data security in the intranet storage system. By the experiment, the MPSFS can provide good security and scalability for intranet massive storage system, and has less effect to the network I/O performance.

A Fourier Shape Descriptor Based on Multi-Level Chord Length Function

A novel Fourier shape descriptor based on multi-level chord length function(MCLFD)is proposed for shape retrieval in this paper.It is derived from equal-arc-length partitions of the shape contour.Compared with the existing Fourier shape descriptors,MCLFD is not only simple to compute but also can reflect the shape's global features and fine details.The MPEG-7 contour shape database and the plant leaf shape database which is taken from nature,are used to test the retrieval performance of MCLFD.The experimental results show that the MCLFD has a very higher retrieval performance than the existing Fourier shape descriptors.The experimental results of adding noises also demonstrate that MCLFD is very robustness.

Investigation of Luhn's claim on information retrieval

In this study, we show how Luhn's claim about the degree of importance of a word in a document can be related to information retrieval. His basic idea is transformed into z-scores as the weights of terms for the purpose of modeling term frequency (tf) within documents. The Luhn-based models represented in this paper are considered as the TF component of proposed TF \times IDF weighing schemes. Moreover, the final term weighting functions appropriate for the TF \times IDF weighting scheme are applied to TREC-6, -7, and -8 databases. The experimental results show relevance to Luhn's claim by having high mean average precision (MAP) for the terms with frequencies around the mean frequency of terms within a document. On the other hand, the weighting, which significantly discriminates the importance between low/high frequencies and medium frequencies, degrades the retrieval performance. Therefore, any weighting scheme (TF) that is directly proportional to tf has a probability of high retrieval performance, if this can optimally indicate the difference of the importance regarding tf values and also optimally eliminate the terms that have high frequencies.

Evaluating the usability of natural language query languages and interfaces to Semantic Web knowledge bases

The need to make the contents of the Semantic Web accessible to end-users becomes increasingly pressing as the amount of information stored in ontology-based knowledge bases steadily increases. Natural language interfaces (NLIs) provide a familiar and convenient means of query access to Semantic Web data for casual end-users. While several studies have shown that NLIs can achieve high retrieval performance as well as domain independence, this paper focuses on usability and investigates if NLIs and natural language query languages are useful from an end-user's point of view. To that end, we introduce four interfaces each allowing a different query language and present a usability study benchmarking these interfaces. The results of the study reveal a clear preference for full natural language query sentences with a limited set of sentence beginnings over keywords or formal query languages. NLIs to ontology-based knowledge bases can, therefore, be considered to be useful for casual or occasional end-users. As such, the overarching contribution is one step towards the theoretical vision of the Semantic Web becoming reality.

Shape alignment and shape orientation analysis-based 3D shape retrieval system

In this paper, we realized an efficient 3D shape retrieval system based on shape alignment and shape orientation analysis. In this system, we suggest extracting the spatial orientation of the polygon surfaces as the feature of one 3D shape. This information is analyzed by multi-resolution wavelet analysis, and the low frequency components are applied to the feature vector. In the preprocessing stage, we adopt four methods of shape alignment including principal component analysis, continuous principal component analysis, normal-based principal component analysis, and plane reflection symmetry analysis (PRSA). We investigated the influence of four alignment methods on retrieval performance. In the orientation sampling stage, the sampling planes are placed on a cube and a dodecahedron, respectively. We also investigated the influence of this two sampling methods on retrieval performance. Finally, one shape descriptor based on plane reflection symmetry analysis and dodecahedron sampling plane (PRSA-DOD) is proposed. We compare this novel shape descriptor, PRSA-DOD descriptor, with the previous methods on Princeton shape benchmark, and results show that this method achieves the higher retrieval performance. This PRSA-DOD descriptor is selected to construct the search engine of system; in the system, the retrieval interface and search engine are implemented.

Evaluating the Usability of Natural Language Query Languages and Interfaces to Semantic Web Knowledge Bases

The need to make the contents of the Semantic Web accessible to end-users becomes increasingly pressing as the amount of information stored in ontology-based knowledge bases steadily increases. Natural language interfaces (NLIs) provide a familiar and convenient means of query access to Semantic Web data for casual end-users. While several studies have shown that NLIs can achieve high retrieval performance as well as domain independence, this paper focuses on usability and investigates if NLIs and natural language query languages are useful from an enduser’s point of view. To that end, we introduce four interfaces each allowing a different query language and present a usability study benchmarking these interfaces. The results of the study reveal a clear preference for full natural language query sentences with a limited set of sentence beginnings over keywords or formal query languages. NLIs to ontology-based knowledge bases can, therefore, be considered to be useful for casual or occasional end-users. As such, the overarching contribution is one step towards the theoretical vision of the Semantic Web becoming reality.

Evaluation of the Texture Analysis Using Spectral Correlation Function

In this paper a new feature extraction technique for texture analysis is evaluated. This method is based on spectral correlation function (SCF) that provides a second-order statistical description in the frequency domain of signals. Two one dimensional signals are obtained from each image by ordering of pixels row by row and column by column. Then the SCF of each signal is calculated by a computational efficient algorithm, namely, the FFT accumulation method (FAM). Features are energy and standard deviation of spectral correlation functions at different regions of bifrequency planes. This scheme shows high performance in the retrieval and classification of Brodatz texture images. Experimental results indicate that the proposedmethod improves retrieval accuracy and correct classification rate in comparing with other approaches. Furthermore the evaluation of the hidden layer output at the classifier with different numbers of neurons indicates that the extracted features from SCF are more separable potentially in comparing with traditional discrete wavelet transform approaches.

One month of human memory consolidation enhances retrieval-related hippocampal activity

We studied the role of the hippocampus in memory retrieval at 1 day and 1 month following associative learning of word pairs. Retrieval-related brain activity was recorded using functional magnetic resonance imaging in 20 healthy students, of which 12 were good learners and eight were poor learners. At the day lag, the poor learners exhibited enhanced neural recruitment in the hippocampus and neocortex to reach a retrieval performance comparable to that of the good learners. Over the 20 subjects, there was a positive correlation between retrieval-related hippocampal activity at the day lag and forgetting over the month retention interval (the greater the activity, the more forgetting). Although the poor learners' retrieval performance declined dramatically from the day to the month lag, the good learners maintained a high retrieval performance, which distinguishes them as good memory consolidators. Their retrieval-related hippocampal and neocortical activity increased from the day to the month lag. This increase was observed both when retrieval performance was matched between the day and the month lag and when the learning procedure for information retrieved at the day and the month lag was matched. This activity increase in the task-specialized neural network from the day lag to the month lag may reflect an increase in task demands or the proliferation of hippocampal-neocortical memory traces during memory consolidation as suggested by the multiple trace theory.

A genetic algorithm-based clustering approach for database partitioning

In a typical distributed/parallel database system, a request mostly accesses a subset of the entire database. It is, therefore, natural to organize commonly accessed data together and to place them on nearby, preferably the same, machine(s)/site(s). For this reason, data partitioning and data allocation are performance critical issues in distributed database application design. We are dealing with data partitioning. Data partitioning requires the use of clustering. Although many clustering algorithms have been proposed, their performance has not been extensively studied. Moreover, the special problem structure in clustering is rarely exploited. We explore the use of a genetic search-based clustering algorithm for data partitioning to achieve high database retrieval performance. By formulating the underlying problem as a traveling salesman problem (TSP), we can take advantage of this particular structure. Three new operators for GAs are also proposed and experimental results indicate that they outperform other operators in solving the TSP. The proposed GA is applied to solve the data-partitioning problem. Our computational study shows that our GA performs well for this application.

An image retrieval system by impression words and specific object names–IRIS

In this paper, we describe a new image retrieval system for scenery images named IRIS (Image Retrieval by Impression words and Specific object names) which uses specific object names in addition to impression words which can reflect ambiguous human kansei (impression and sensitivity) as the retrieval keywords. IRIS has not only the function which retrieves images but also that can estimate keywords to images automatically. In IRIS, first an image is divided into some regions. Next each region is roughly classified into the sky, earth and water categories by the image recognition method using a neural network. Then the image characteristics are extracted from each category, and the impression words are given to the image automatically. After the regions are classified into sky/earth/water categories, they are classified into much more detailed objects such as mountain or cloudy weather sky. By the hierarchical recognition, the specific object names are given to an image automatically. Using concrete object names with vague impression words as keywords, flexible retrieval is possible. We carried out extensive experiments to test the performance of the developed IRIS. The retrieval experiments has shown the efficiency of retrieval when the impression words and the specific object names are used together. Also, they have shown the high retrieval performances of IRIS.

An effective World Wide Web image search engine

Effective World Wide Web image retrieval systems are required to locate relevant images as more and more images are used in HTML documents. There are two general approaches to image retrieval: one is based on text descriptions and the other on image contents (low-level features, such as colour and texture). They have their respective strengths and weaknesses. In this paper, we describe how to improve text-based and content-based image retrieval techniques and to integrate these two types of techniques in one system to take advantage of their complementing strengths. Our experimental results show that the integrated approach has higher retrieval performance than either the text-based or the content-based techniques used by themselves.

Color image retrieval using hybrid graph representation

In this paper, a robust color image retrieval algorithm is proposed based on the hybrid graph representation, i.e., a dual graph which consists of the Modified Color Adjacency Graph (MCAG) and Spatial Variance Graph (SVG). The MCAG, which is similar to the Color Adjacency Graph (CAG) [6], is proposed to enhance the indexing ability and the database capacity, by increasing the feature dimension. In addition, the SVG is introduced, in order to utilize the geometric statistics of the chromatic segment in the spatial domain. In the matching process, we expand the histogram intersection [2] into the graph intersection, in which graph matching is performed using simple matrix operations. Intensive discussions and experimental results are provided to evaluate the performance of the proposed algorithm. Experiments are carried out on the Swain's test images and the Virage images, demonstrating that the proposed algorithm yields high retrieval performance with tolerable computational complexity. It is also shown that the proposed algorithm works well, even if the query image is corrupted. e.g., a large part of pixels is missing.

Communication and Computing for Distributed Multimedia Systems

i»?i»?Guojun Lu Artech House, Cambridge, MA, 1996, 416 pp. ISBN 0-89006-884-4, $69.00 The telecommunications and computer industries are experiencing explosive growth. The most important trend is the convergence of the two in pursuing the promise of the information superhighway, a high-speed, ubiquitous, and seamless integrated services information infrastructure. The book Communication and Computing for Distributed Multimedia Systems by Lu focuses on a converged approach to development of distributed multimedia systems to meet the stringent requirements of multimedia computing and communications applications. It provides the first systematic discussion of issues and technology in developing networked multimedia systems. Each chapter covers issues related to an aspect of distributed multimedia systems and concludes with a summary, problems, and references. These help the readers better understand the text and provide insightful research directions for further investigations. The new concepts in each chapter are clearly defined, which clarifies the confusion caused by different new terms recently used in the literature. Chapter 1 introduces the fundamental concepts of multimedia computing and communications with sample applications. It also describes the different types of multimedia systems and the challenges of developing distributed multimedia systems. Chapter 2 presents the characteristics of multimedia data including digital audio, image, and video data. It then discusses the system requirements of processing multimedia data. It also includes a unique explanation of color specification and its role in achieving high picture quality, high compression ratio, and high information retrieval performance critical for future multimedia systems. Chapter 3 is devoted to the discussion of the principles and current techniques and standards for digital audio, image, and video compression which are very important, since multimedia data require a large storage to store and a large bandwidth to transmit. The challenge in developing distributed multimedia systems is to provide performance guarantees while using the network resources efficiently. Chapter 4 presents the end-to-end quality-of-service (QoS) framework for continuous media communication. In the next four chapters, the book discusses the support of QoS guarantee required at different layers of the communications network. Chapter 5 examines the characteristics of networks suitable for multimedia communication and discusses technical issues of QoS guarantees at the network layer. Chapter 6 covers the design issues of multimedia transport protocols and reviews a number of proposed multimedia transport protocols for the transport layer. Chapter 7 discusses the end-system support, including the support of hardware and operating systems, which is as important as network and transport support for multimedia communication. Chapter 8 is devoted to the discussion of a special type of end systems, multimedia servers. The book also covers techniques for networked multimedia synchronization and multimedia information indexing and retrieval. Chapter 9 discusses the requirements and mechanisms for intramedia and intermedia synchronization of multimedia data streams in distributed multimedia systems. Chapter 10 addresses how to organize multimedia information so that relevant information can be retrieved quickly from multimedia servers and techniques for indexing and retrieving multimedia information in distributed multimedia systems. The last chapter, Chapter 11, describes four current distributed multimedia systems: the World Wide Web (WWW), the Multicast Backbone (MBone), Video-on-Demand (VoD), and videoconferencing. The four systems can be considered as the primitive parts of the information superhighway. It concludes with a discussion of the general requirements and components of the information superhighway. From this book, the readers can learn about the special demands multimedia systems place on the computer architecture, storage, file system, operating system, and communications systems, and better understand the complex design challenges these components present. In addition, the readers can gain insight into current techniques and research efforts aimed at developing new computer architecture, operating systems, and communications systems to support multimedia. Looking ahead, the readers can catch a futuristic glimpse at the common networking paradigm evolved from the current telecommunications and computer networking models. Although the book has a relatively complete and balanced coverage of technical issues of distributed multimedia systems with QoS guarantees, it does not discuss two important issues that must be considered in development of distributed multimedia systems, security and user-friendly access. Hongchi Shi, University of Missouri-Columbia

Xtract: An overview

Lexical collocations have particular statistical distributions. We have developed a set of statistical techniques for retrieving and identifying collocations from large textual corpora. The techniques we developed are able to identify collocations of arbitrary length as well as flexible collocations. These techniques have been implemented in a lexicographic tool, Xtract, which is able to automatically acquire collocations with high retrieval performance. Xtract works in three stages. The first stage is based on a statistical technique for identifying word pairs involved in a syntactic relation. The words can appear in the text in any order and can be separated by an arbitrary number of other words. The second stage is based on a technique to extract n-word collocations (or n-grams) in a much simpler way than related methods. These collocations can involve closed class words such as particles and prepositions. A third stage is then applied to the output of stage one and applies parsing techniques to sentences involving a given word pair in order to identify the proper syntactic relation between the two words. A secondary effect of the third stage is to filter out a number of candidate collocations as irrelevant and thus produce higher quality output. In this paper we present an overview of Xtract and we describe several uses for Xtract and the knowledge it retrieves such as language generation and machine translation.