Search

In this paper, we present a multiple representations and multiple queries approach to tackle the problem of invariance in the framework of content-based image retrieval (CBIR), especially in the case of texture. This approach, rather than considering invariance at the representation level, considers it at the query level. We use two models to represent texture visual content, namely the autoregressive model and a perceptual model based on a set of perceptual features. The perceptual model is used with two viewpoints: the original images viewpoint and the autocovariance function viewpoint. After a brief presentation and discussion of these multiple representation models / viewpoints, which are not invariant with respect to geometric and photometric transformations, we present the invariant texture retrieval algorithm, which is based on multiple models / viewpoints and multiple queries approach and consists in two levels of results fusion (merging): 1. The first level consists in merging results returned by the different models / viewpoints (representations) for the same query in one results list using a linear results fusion model; 2. The second level consists in merging each fused list of different queries into a unique fused list using a round robin fusion scheme. Experimentations show promising results. [ABSTRACT FROM AUTHOR]

In this paper we propose to revisit the well-known autoregressive model (AR) as a texture representation model. We consider the AR model with causal neighborhoods. First, we will define the AR model and discuss briefly the parameters estimation process. Then, we will present the synthesis algorithm and we will show some experimental results. The causal autoregressive model is applied in content-based image retrieval. Benchmarking conducted on the well-known Brodatz database shows interesting results. Both retrieval effectiveness (relevance) and retrieval efficiency are discussed and compared to the well-known multiresolution simultaneous autoregressive model (MRSAR). [ABSTRACT FROM AUTHOR]

In this paper we introduce the design, implementation and evaluation of the Dynamic Visualization Toolkit (DyVT) to support complex dynamic social network visualization. Dynamic aspects of social networks such as spatiotemporal as well as personalized information can be visualized in a common toolkit. To that end, an XML-based target language DyVTML is an extension of existing schemata enabling expression, storage and interchange of rich animated social network data. With the language and the available tool support, even less-experienced users can visualize temporal data in animations and spatial data in maps and personalize it with icons and colors. The prototype is evaluated by the visualization of large mailing list data sets. [ABSTRACT FROM AUTHOR]

In this paper we propose a method for detection of the car license plates in 2D gray images. In this method we first estimate the density of vertical edges in the image. The regions with high density vertical edges are good candidates for license plates. In order to filter out clutter regions possessing similar feature in the edge density image, we design a match filter which models the license plate pattern. By applying the proposed filter on the edge density image followed by a thresholding procedure, the locations of license plate candidates are detected. We finally extract the boundary of license plate(s) using the morphological operations. The result of experiments on car images (taken under different imaging conditions especially complex scenes) confirms the ability of the method for license plate detection. As the complexity of the proposed algorithm is low, it is considerably fast. [ABSTRACT FROM AUTHOR]

In this paper, we describe a visual language, based on the so-called Predicate Tree Metaphor, which allows users to visually build complex sentences for querying commonly used search engines. By using this visual language, no parentheses have to be applied, and no precedence rules have to be known. Promising results about the usability of the proposed interface are reported, on the basis on an experimental between-group study, performed on a Yahoo-based prototype of the proposed graphical environment. [ABSTRACT FROM AUTHOR]

Chorems are schematized representations of territories, and so they can represent a good visual summary of spatial databases. Indeed for spatial decision-makers, it is more important to identify and map problems than facts. Until now, chorems were made manually by geographers based on the own knowledge of the territory. So, an international project was launched in order to automatically discover spatial patterns and layout chorems starting from spatial databases. After examining some manually-made chorems some guidelines were identified. Then the architecture of a prototype system is presented based on a canonical database structure, a subsystem for spatial patterns discovery based on spatial data mining, a subsystem for chorem layout, and a specialized language to represent chorems. [ABSTRACT FROM AUTHOR]

In our project on the design of intelligent applications for an office environment we focus on the coffee corner as it is the place in an office where professional and social interactions intermingle. In this article the user-centered design of tangible interfaces, i.e. i-Candies and i-Bowl, which have been used as input devices to access company information in a coffee corner, is described. The i-Candies and the i-Bowl illustrate how everyday objects can be part of a context-aware system. By picking up one of the i-Candies and placing it on the i-Bowl, the i-Candy controls which information is presented on the wall display in the coffee corner. In our evaluation, we found that the office workers not only were enticed to interact with the i-Candies, but also that the i-Candies provided mediums that stimulate sociability among office workers. [ABSTRACT FROM AUTHOR]

Geometric modelling and haptic rendering of textiles is an area of research in which interest has significantly increased over the last decade. A haptic representation is created by adding the physical properties of an object to its geometric configuration. While research has been conducted into geometric modelling of fabrics, current systems require textile data to be manually entered into the computer simulation by a technician. This study explores the possibility of automatic generation of geometric and haptic models of real world textile samples. The development of a scalable and generic methodology for geometric and haptic modelling of plain weave textiles made from wool yarn is reported. This system has been successfully implemented using a step-wise procedure. Initially, an image of the textile artefact is captured. Then the critical features of the image are extracted from the image and deployed in a finite element model. The geometric model is augmented by adding physical properties of the textile and developing the haptic model. Two different haptic rendering procedures are implemented based on Reachin Application Programming Interface 3.2 (API). The developed methodologies are described and results obtained are provided. [ABSTRACT FROM AUTHOR]

In digital holography a CCD camera records optically generated holograms which is then reconstructed numerically by a calculation of scalar diffraction in the Fresnel approximation. The digital photography facilitates real time transmission of the message via traditional communication methods. In this paper the principle of digital holography and its application to the 3D image encryption-decryption are reviewed. The experimental results of firearm identification recording using digital holography and their numerical reconstruction are presented. [ABSTRACT FROM AUTHOR]

In this paper, we introduce a novel high-level visual content descriptor devised for performing semantic-based image classification and retrieval. The work can be treated as an attempt for bridging the so called "semantic gap". The proposed image feature vector model is fundamentally underpinned by an automatic image labelling framework, called Collaterally Cued Labelling (CCL), which incorporates the collateral knowledge extracted from the collateral texts accompanying the images with the state-of-the-art low-level visual feature extraction techniques for automatically assigning textual keywords to image regions. A subset of the Corel image collection was used for evaluating the proposed method. The experimental results indicate that our semantic-level visual content descriptors outperform both conventional visual and textual image feature models. [ABSTRACT FROM AUTHOR]

In response to the need of using electronic design data directly in construction management applications, many CAD developers have started implementing semantic data models in their CAD products using industry foundation classes (IFCs). While helpful, these semantic CAD applications have limitations when used in actual construction practices. The case studies in the thesis indicated that: 1) the semantics of the current data model (IFC as an example) is not rich enough to cover the richer details of the real trade practices; 2) the current implementation of the semantic data model lacks the mechanism to provide multiple trades views at various detailed levels. This paper also provided suggestions for the future development of semantic data model of construction industry. [ABSTRACT FROM AUTHOR]

A Neurovision System can be defined as an artificial tool that sees our physical world. The purpose of this paper is to show a novel tool to design a 3D artificial vision system based on Resilient Neural Networks. Camera Calibration (CC) is a fundamental issue for Computational-Vision. Classical CC methods comprise of taking images of objects with known geometry, extracting the features of the objects from the images, and minimizing their 3D backprojection errors. In this paper, a novel implicit-CC model based on Resilient Neural Networks, CR, has been introduced. The CR is particularly useful for 3D reconstruction of the applications that do not require explicitly computation of physical camera parameters in addition to the expert knowledge. The CR supports intelligent-photogrammetry, photogrammetron. In order to evaluate the success of the proposed implicit-CC model, the 3D reconstruction performance of the CR has been compared with two different well-known implementations of the Direct Linear Transformation (DLT). The proposed method is also robust sufficiently for dealing with different cameras because it is capable of fusion of the image coordinates sourced from different cameras once the neural network has been trained. [ABSTRACT FROM AUTHOR]

Just-in-Time Information Retrieval (JITIR) systems aim to automatically retrieve useful information on the basis of the user's current task and to present this information to the user without disrupting that task. We ask whether the cognitive mechanism of ‘selective semantic processing' can minimise the disruptive nature of presenting JITIR information to the user. This mechanism may allow users to subconsciously filter out irrelevant information presented in the periphery of the visual field, while maintaining awareness of relevant information. We report an experiment assessing both attention to peripherally presented information (measured via recall) and level of distraction (measured via typing keystroke rate) in a JITIR system used to write reports on various topics. The experimental results showed that peripherally presented information that was relevant to a user's writing topic reliably entered their attention significantly more often than irrelevant information, and was significantly less distracting than similar but irrelevant information. [ABSTRACT FROM AUTHOR]

A geographic map is an important browsing tool for multimedia data that can include personal photos, but geographically correct maps are not always easy to use for that purpose due to the frequent zooming and panning, as well as the existence of extraneous information. This paper proposes a new user-interface concept for geo-tagged personal multimedia browsing in the form of a cognitive map. In addition, design criteria are defined and an auto-generation method is presented for this map. The proposed method produces a map represented as a clustered graph with vertices and edges in real time. It is visually compact, preserves geographical relationships among locations and is designed for both PCs and mobile devices. An experiment was conducted to test the proposed method with real-life data sets. [ABSTRACT FROM AUTHOR]

Automatic image annotation aims at labeling images with keywords. In this paper we investigate three annotation benchmark tasks used in literature to evaluate annotation systems' performance. We empirically compare the first two of the tasks, the 5000 Corel images and the Corel categories tasks, by applying a family of annotation system configurations derived from our PicSOM image content analysis framework. We establish an empirical correspondence of performance levels in the tasks by studying the performance of our system configurations, along with figures presented in literature. We also consider ImageCLEF 2006 Object Annotation Task that has earlier been found difficult. By experimenting with the data, we gain insight into the reasons that make the ImageCLEF task difficult. In the course of our experiments, we demonstrate that in these three tasks the PicSOM system—based on fusion of numerous global image features—outperforms the other considered annotation methods. [ABSTRACT FROM AUTHOR]

The detection of commercials in news video has been a challenging problem because of the diversity of the production styles of commercial programs. In this paper, the authors present a novel algorithm for the detection of commercials in news program. By the method suggested, firstly shot transition detection and anchorman shot recognition are conducted, then clustering analysis is employed to label commercial blocks roughly, finally the accurate boundaries of the commercials are located by analyzing the average duration of preceding and subsequent shots and the visual features of the shots, such as color, saturation and edge distribution. The experiment results show that the proposed algorithm is effective with high precision. [ABSTRACT FROM AUTHOR]

In this study, we propose a simple, yet general and powerful framework for constructing accurate affine invariant regions and use it for near-duplicate detection problem. In our framework, a method for extracting reliable seed points is first proposed. Then, regions which are invariant to most common affine transformations are extracted from seed points by a new method named the Thresholding Seeded Growing Region (TSGR). After that, an improved ellipse fitting method based on the Direct Least Square Fitting (DLSF) is used to fit the irregularly-shaped contours of TSGRs to obtain ellipse regions as the final invariant regions. At last, SIFT-PCA descriptors are computed on the obtained regions. In the experiment, our framework is evaluated by retrieving near-duplicate in an image database containing 1000 images. It gives a satisfying result of 96.8% precision at 100% recall. [ABSTRACT FROM AUTHOR]

Face image can be seen as a complex visual object, which combines a set of characterizing facial features. These facial features are crucial hints for machine to distinguish different face images. However, the face image also contains certain amount of redundant information which can not contribute to the face image retrieval task. Therefore, in this paper we propose a retrieval system which is aim to eliminate such effect at three different levels. The Ternary Feature Vector (TFV) is generated from quantized block transform coefficients. Histograms based on TFV are formed from certain subimages. Through this way, irrelevant information is gradually removed, and the structural and statistical information are combined. We testified our ideas over the public face database FERET with the Cumulative Match Score evaluation. We show that proper selection of subimage and feature vectors can significantly improve the performance with minimized complexity. Despite of the simplicity, the proposed measures provide results which are on par with best results using other methods. [ABSTRACT FROM AUTHOR]

The paper presents a framework for information retrieval in visual databases containing colour images. The concept of approximation-based keypoints is adapted to colour images; building and detection of such keypoints are explained in details. The issues of matching images are only briefly highlighted. Finally, the idea of higher-level keypoints is proposed. [ABSTRACT FROM AUTHOR]

Image similarity is an important topic in the field of image processing. The goal is to obtain objective measures that express the similarity between two images in a way that matches human evaluation. Such measures have both theoretical and practical applications. In this paper, we show how similarity measures for fuzzy sets have been modified in order to be applied in image processing. We also discuss a new application of these measures in the context of color image retrieval, indicating the potential of this class of similarity measures. [ABSTRACT FROM AUTHOR]

Stochastic modulation steganography hides secret message within the cover image by adding a weak noise signal with a specified probabilistic distribution. The advantages of stochastic modulation steganography include high capacity and better security. Current steganalysis methods that are applicable to the detection of hidden message in traditional least significant bit (LSB) or additive noise model based steganography cannot reliably detect the existence of hidden message in stochastic modulation steganography. In this paper, we present a new steganalysis approach which can reliably detect the existence and accurately estimate the length of hidden message in stochastic modulation steganography. By analyzing the distributions of the horizontal pixel difference of the images before and after stochastic modulation embedding, it is shown that for non-adaptive steganography, the distribution of the stego-image's pixel difference can be modeled as the convolution of the distribution of the cover image's pixel difference and that of the quantized stego-noise difference, and that the estimation of the hidden message length in stochastic modulation can be achieved by estimating the variance of the stego-noise. To estimate the variance of the stego-noise, hence determining the existence and the length of hidden message, we first model the distribution of the cover image's pixel difference as a generalized Gaussian and estimate the parameters of the distribution using grid search and Chi-square goodness of fit test, and then exploit the relationship between the distribution variance of the cover image's pixel difference and that of the stego-noise difference. We present experimental results to demonstrate that our new approach is effective for steganalyzing stochastic modulation steganography. Our method provides a general theoretical framework and is applicable to other non-adaptive embedding algorithms where the distribution models of the stego-noise are known or can be estimated. [ABSTRACT FROM AUTHOR]