De Xu

Beijing Jiaotong University, Peping, Beijing, China

Are you De Xu?

Claim your profile

Publications (103)33.82 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: A DataSpace Support Platform (DSSP) is a self-sustained and self-managed system which needs to support uncertainty among its mediated schemas and its schema mappings. Some approaches for managing such uncertainty by assigning probabilities and reliability degrees to schema mappings have been proposed. Unfortunately, the number of mappings self-generated by a DSSP is usually too large and among those possible mappings, some might be totally correct and others partially correct. Therefore, providing probabilities or reliability degrees to the mappings is necessary but not sufficient to resolve uncertainty among them. This paper proposes a stepper-based approach called pos-mapping to managing reliable mappings using possibility theory. Instead of choosing a threshold for managing the reliable mappings, pos-mapping approach orders and divides the set of reliable mappings into subsets of possibility distributions and assigns to each of these subsets a recursive possibility degree function. The recursiveness of the possibility degree function leads to an incremental management of the possibility distributions. Experimental results show that our system is more efficient than the existing systems and the accuracy of the results increases with the number of reliable schemas in the DSSP.
    No preview · Article · May 2014 · International Journal of Software Engineering and Knowledge Engineering
  • [Show abstract] [Hide abstract]
    ABSTRACT: Contrary to existing heterogeneous data integration systems which need to be fully integrated before using, a Dataspace Support Platform is a self-sustained system which automatically provides for the user its best endeavor results regardless of how integrated its sources are. Therefore, a Dataspace Support Platform needs to support uncertainty in mediated schema and in schema mappings. This paper proposes a novel approach to automatically providing reliable mediated schemas and reliable semantic mappings in Dataspace Support Platforms. Our aim is to increase the system's endeavor results by leading it to considering as much as possible information available in any source connected. In fact, we first extract from the source schemas, their corresponding graph representations. Then, we introduce algorithms which automatically extract a set of mediated schemas from the graph representations and a set of semantic mappings between a source and a target mediated schema. Finally, we assign reliability degrees to the mediated schema generated and to the semantic mappings. Indeed, the higher the reliability degree of a given mediated schema or semantic mapping, the more consistent with the source it is. Compared with existing systems, experimental results show that our system is faster and, although completely automatic, it produces reliable mediated schemas and reliable semantic mappings which are as accurate as those produced by semi-automatic systems.
    No preview · Article · Jan 2013 · Computing and Informatics
  • [Show abstract] [Hide abstract]
    ABSTRACT: Combinatorial maps are widely used in image representation and processing, however map matching problems have not been extensively researched. This paper addresses the problem of inexact matching between labeled combinatorial maps. First, the concept of edit distance is extended to combinatorial maps, and then used to define mapping between combinatorial maps as a sequence of edit operations that transforms one map into another. Subsequently, an optimal approach based on A* algorithm and an approximate approach based on Greedy algorithm are proposed to compute the distance between combinatorial maps. Experimental results show that the proposed inexact map matching approach produces richer search results than the exact map matching technique by tolerating small difference between maps. The proposed approach performs better in practice than the previous approach based on maximum common submap which cannot be directly used for comparing labels on the maps.
    No preview · Article · Dec 2012 · Computer Vision and Image Understanding
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces 2PROM, a new algorithm that can efficiently retrieve information from a set of multimedia and heterogeneous data sources. We published IHM, a model to predict whether a x-ray picture carries the trace of cancer (viruses). We further generalized our approach to The NoCancerSpace, a dataspace for medical diagnosis of lung cancer. This paper presents the optimization algorithm used by the NocancerSpace. 2PROM is designed to optimize the Dataspace retrieval process with two main phases. Its first phase consists of building a pipeline to find the best retrieval strategies. In fact, the pipeline explores the set of alternative execution strategies to determine the cheapest one. The retrieval strategies are initial nodes of the next phase. As for the second phase, retrieval strategies are combined with predictive model to determine the most efficient way to execute a query. In order words, the optimizer considers the possible retrieval strategies for a given input query, and attempts to determine which of those strategies will be the most efficient. The retrieval strategies are represented as XML tree of “strategy nodes”. The output of the second phase is the best results found. Experiments show that 2PROM retrieves more relevant results in less time than existing systems.
    No preview · Conference Paper · Oct 2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Iconic communication is paramount today in order to assist people with disability (e.g. illiteracy) enjoying, as much as everybody else, the advances in information and communication technologies (e.g. Internet). Previous works tend to generalize iconic communication by translating iconic sentences into XML documents. Theses approaches are limited owing to the fact that an icon can hide several metaphors. In fact, the semantics of an icon is not the linguistic equivalent associated to the image, but is a set of attributes which can be used to describe the given icon. Second, an XML schema is not a knowledge representation, but just a message format. Therefore, to manage the knowledge hidden behind iconic sentences, a semantic model for icons needs to be formally defined. This paper extends previous icon models by first, introducing a description logics-based definition of icons semantics, and second, based on those formal definitions and the Web Ontology Language (OWL), we create an Ontology for Icons named IcOnto (read “eye can too”). We further use IcOnto to model some properties of the African Traditional Medicine (ATM), for illustration.
    No preview · Conference Paper · Oct 2012
  • Ning Wang · Congyan Lang · De Xu
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of illumination estimation is to measure the illuminant color of an image. Various methods have been proposed to handle this problem. Recent Gray Edge algorithm is proved to be an effective one. It uses the edge information in the image to estimate the illumination color. In the original literature of Gray Edge, gaussian filter is incorporated to remove the edge noises. But there is a serious problem with gaussian filter, the weighted value of the gaussian influence function only depends on the spatial distance between the pixels and completely ignores their value. As a result, although noises are removed, the physical edges are also blurred. So the accuracy of illumination estimation will be influenced. In this paper, instead of gaussian filter, we incorporated bilateral filter and the Gray Edge algorithm together. Bilateral filter is a non-linear technique that can blur an image while respecting strong edges. With more accurate edge information, better results could be expected for the Gray Edge algorithm. And the experimental results proved our deduction.
    No preview · Article · May 2012 · Journal of Computational and Theoretical Nanoscience
  • Hong Bao · De Xu · Yingjun Tang
    [Show abstract] [Hide abstract]
    ABSTRACT: Visually saliency detection provides an alternative methodology to image description in many applications such as adaptive content delivery and image retrieval. One of the main aims of visual attention in computer vision is to detect and segment the salient regions in an image. In this paper, we employ matrix decomposition to detect salient object in nature images. To efficiently eliminate high contrast noise regions in the background, we integrate global context information into saliency detection. Therefore, the most salient region can be easily selected as the one which is globally most isolated. The proposed approach intrinsically provides an alternative methodology to model attention with low implementation complexity. Experiments show that our approach achieves much better performance than that from the existing state-of-art methods.
    No preview · Article · May 2012 · IEICE Transactions on Information and Systems
  • Source
    Hongzhe Liu · Hong Bao · De Xu
    [Show abstract] [Hide abstract]
    ABSTRACT: We define WordNet based hierarchy concept tree (HCT) and hierarchy concept graph (HCG), HCT contains hyponym/hypernym kind of relation in WordNet while HCG has more meronym/holonym kind of edges than in HCT, and present an advanced concept vector model for generalizing standard representations of concept similarity in terms of WordNet-based HCT. In this model, each concept node in the hierarchical tree has ancestor and descendent concept nodes composing its relevancy nodes, thus a concept node is represented as a concept vector according to its relevancy nodes’ local density and the similarity of the two concepts is obtained by computing the cosine similarity of their vectors. In addition, the model is adjustable in terms of multiple descendent concept nodes. This paper also provides a method by which this concept vector may be applied with regard to HCG into HCT. With this model, semantic similarity and relatedness are computed based on HCT and HCG. The model contains structural information inherent to and hidden in the HCT and HCG. Our experiments showed that this model compares favorably to others and is flexible in that it can make comparisons between any two concepts in a WordNet-like structure without relying on any additional dictionary or corpus information.
    Full-text · Article · Feb 2012 · Journal of Systems and Software
  • Ning Wang · Congyan Lang · De Xu
    [Show abstract] [Hide abstract]
    ABSTRACT: Removing shadows from single color images is an important problem in computer vision. In this paper, we propose a novel shadow removal approach, which could effectively remove shadows from textured surfaces, yielding high quality shadow-free images. Our approach aims at calculating scale factors to cancel the effect of shadows. Based on the regional gray edge hypothesis, which assumes the average of the reflectance differences in a region is achromatic, the scale factors can be computed without the restrictions that former algorithms need. The experimental results show that the proposed algorithm is effective.
    No preview · Article · Dec 2011 · Optical Engineering
  • Ning Wang · Congyan Lang · De Xu
    [Show abstract] [Hide abstract]
    ABSTRACT: Removing shadows in color images is an important research problem in computer vision. In this paper, we propose a novel shadow removal approach, which effectively removes shadows from textured surfaces, yielding high quality shadow-free images. Our approach aims at calculating scale factors to cancel the effect of shadows. Based on the regional gray edge hypothesis, which assumes the average of the reflectance differences in a region is achromatic, the scale factors can be computed without the restrictions that former algorithms need. The experimental results show that the proposed algorithm is effective and improves the performance of former scale-based shadow removal methods.
    No preview · Article · Nov 2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: Tag ranking has emerged as an important research topic recently due to its potential application on web image search. Existing tag relevance ranking approaches mainly rank the tags according to their relevance levels with respect to a given image. Nonetheless, such algorithms heavily rely on the large-scale image dataset and the proper similarity measurement to retrieve semantic relevant images with multi-labels. In contrast to the existing tag relevance ranking algorithms, in this paper, we propose a novel tag saliency ranking scheme, which aims to automatically rank the tags associated with a given image according to their saliency to the image content. To this end, this paper presents an integrated framework for tag saliency ranking, which combines both visual attention model and multi-instance learning to investigate the saliency ranking order information of tags with respect to the given image. Specifically, tags annotated on the image-level are propagated to the region-level via an efficient multi-instance learning algorithm firstly; then, visual attention model is employed to measure the importance of regions in the given image. Finally, tags are ranked according to the saliency values of the corresponding regions. Experiments conducted on the COREL and MSRC image datasets demonstrate the effectiveness and efficiency of the proposed framework.
    No preview · Article · Oct 2011 · Neurocomputing
  • Shuoyan Liu · De Xu · Songhe Feng
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes a method for scene categorization by integrating region contextual information into the popular Bag-of-Visual-Words approach. The Bag-of-Visual-Words approach describes an image as a bag of discrete visual words, where the frequency distributions of these words are used for image categorization. However, the traditional visual words suffer from the problem when faced these patches with similar appearances but distinct semantic concepts. The drawback stems from the independently construction each visual word. This paper introduces Region-Conditional Random Fields model to learn each visual word depending on the rest of the visual words in the same region. Comparison with the traditional Conditional Random Fields model, there are two areas of novelty. First, the initial label of each patch is automatically defined based on its visual feature rather than manually labeling with semantic labels. Furthermore, the novel potential function is built under the region contextual constraint. The experimental results on the three well-known datasets show that Region Contextual Visual Words indeed improves categorization performance compared to traditional visual words.
    No preview · Article · Sep 2011 · Expert Systems with Applications
  • Tao Wang · Guojun Dai · De Xu
    [Show abstract] [Hide abstract]
    ABSTRACT: Combinatorial maps explicitly encode orientations of edges around vertices, and have been used in many fields. In this paper, we address the problem of searching for patterns in model maps by putting forward the concept of symbol graph. A symbol graph will be constructed and stored for each model map in the preprocessing. Furthermore, an algorithm for submap isomorphism is presented based on symbol sequence searching in the symbol graphs. The computational complexity of this algorithm is quadratic in the worst case if we neglect the preprocessing step.
    No preview · Article · Jun 2011 · Pattern Recognition Letters
  • [Show abstract] [Hide abstract]
    ABSTRACT: Localized content-based image retrieval (LCBIR) has emerged as a hot topic more recently because in the scenario of CBIR, the user is interested in a portion of the image and the rest of the image is irrelevant. In this paper, we propose a novel region-level relevance feedback method to solve the LCBIR problem. Firstly, the visual attention model is employed to measure the regional saliency of each image in the feedback image set provided by the user. Secondly, the regions in the image set are constructed to form an affinity matrix and a novel propagation energy function is defined which takes both low-level visual features and regional significance into consideration. After the iteration, regions in the positive images with high confident scores are selected as the candidate query set to conduct the next-round retrieval task until the retrieval results are satisfactory. Experimental results conducted on the SIVAL dataset demonstrate the effectiveness of the proposed approach.
    No preview · Article · Jun 2011 · IEICE Transactions on Information and Systems
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents an efficient yet powerful codebook model, named classified codebook model, to categorize natural scene category. The current codebook model typically resorts to large codebook to obtain higher performance for scene categorization, which severely limits the practical applicability of the model. Our model formulates the codebook model with the theory of vector quantization, and thus uses the famous technique of classified vector quantization for scene-category modeling. The significant feature in our model is that it is beneficial for scene categorization, especially at small codebook size, while saving much computation complexity for quantization. We evaluate the proposed model on a well-known challenging scene dataset: 15 Natural Scenes. The experiments have demonstrated that our model can decrease the computation time for codebook generation. What is more, our model can get better performance for scene categorization, and the gain of performance becomes more pronounced at small codebook size.
    No preview · Article · Jun 2011 · IEICE Transactions on Information and Systems
  • [Show abstract] [Hide abstract]
    ABSTRACT: One possible solution to estimating the illumination for color constancy and white balance in video sequences would be to apply one of the many existing illumination-estimation algorithms independently to each video frame. However, the frames in a video are generally highly correlated, so we propose a video-based illumination-estimation algorithm that takes advantage of the related information between adjacent frames. The main idea of the method is to cut the video clip into different ‘scenes.’ Assuming all the frames in one scene are under the same (or similar) illuminant, we combine the information from them to calculate the chromaticity of the scene illumination. The experimental results showed that the proposed method is effective and outperforms the original single-frame methods on which it is based.
    No preview · Conference Paper · Apr 2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: Until now, most exposure fusion methods are easy to be influenced by the location of object in the image. However, when capturing the source images, slight shift in the camera's position will yield blurry or double images. In order to solve the problem, a method called SIDWTBEF is proposed, which is based on shift-invariant discrete wavelet transform (SIDWT). It is more robust to images those have slight shift. On the other hand, in this paper, we present a novel way to get the chrominance information of the scene, and the saturation of the fused image can be adjusted using one user-controlled parameter. The luminance images sequence of the source images are decomposed by SIDWT into subimages with a certain level scale. In the transform domain, different fusion rules are used for the high-pass sub-images and the low-pass sub-images combination respectively. In the end, in order to reduce the inconsistencies induced by the fusion rule after applying the inverse transform of SIDWT, an enhancement operator is proposed. Experiments show that SIDWTBEF can give comparative results compared to other shift dependent exposure fusion methods.
    No preview · Article · Jan 2011 · Journal of Information Science and Engineering
  • [Show abstract] [Hide abstract]
    ABSTRACT: Previous works show that the probabilistic Latent Semantic Analysis (pLSA) model is one of the best generative models for scene categorization and can obtain an acceptable classification accuracy. However, this method uses a certain number of topics to construct the final image representation. In such a way, it restricts the image description to one level of visual detail and cannot generate a higher accuracy rate. In order to solve this problem, we propose a novel generative model, which is referred to as multi-scale multi-level probabilistic Latent Semantic Analysis model (msml-pLSA). This method consists of two parts: multi-scale part, which extracts visual details from the image of diverse resolutions, and multi-level part, which concentrates multiple levels of topic representation to model scene. The msml-pLSA model allows for the description of fine and coarse local image detail in one framework. The proposed method is evaluated on the well-known scene classification dataset with 15 scene categories, and experimental results show that the proposed msml-pLSA model can improve the classification accuracy compared with the typical classification methods.
    No preview · Article · Jan 2011 · IEICE Transactions on Information and Systems
  • Shuoyan Liu · De Xu · Songhe Feng
    [Show abstract] [Hide abstract]
    ABSTRACT: Emotion categorization of natural scene images represents a very useful task for automatic image analysis systems. Psychological experiments have shown that visual information at the emotion level is aggregated according to a set of rules. Hence, we attempt to discover the emotion descriptors based on the composition of visual word representation. First, the composition of visual word representation models each image as a matrix, where elements record the correlations of pairwise visual words. In this way, an image collection is modeled as a third-order tensor. Then we discover the emotion descriptors using a novel affective-probabilistic latent semantic analysis (affective-pLSA) model, which is an extension of the pLSA model, on this tensor representation. Considering that the natural scene image may evoke multiple emotional feelings, emotion categorization is carried out using the multilabel k-nearest-neighbor approach based on emotion descriptors. The proposed approach has been tested on the International Affective Picture System and a collection of social images from the Flickr website. The experimental results have demonstrated the effectiveness of the proposed method for eliciting image emotions.
    No preview · Article · Dec 2010 · Optical Engineering
  • Hong Bao · Ye Liang · Hong-Zhe Liu · De Xu
    [Show abstract] [Hide abstract]
    ABSTRACT: Traditional Chinese painting is a unique form of art and highly regarded for its theory, expression and techniques throughout the world. A traditional Chinese painting is composed of three parts: the mainbody part, the seals part and scripts part. These three parts have many semantics. So extraction of them is important and urgent task. However, popular image processing techniques have little been used in this specific domain. In this paper, a novel algorithm for extraction the scripts part of traditional Chinese painting images is proposed, including the motivations of the algorithm, the description of the algorithm, experiment results and its analysis. This algorithm is mainly based on color and structure feature of Chinese characters in the scripts part of traditional Chinese painting images. The algorithm is simple but has satisfactory efficiency.
    No preview · Conference Paper · Nov 2010