Showing papers by "Xi'an Jiaotong University published in 2002"

Stereo Matching Using Belief Propagation

Abstract: In this paper, we formulate the stereo matching problem as a Markov network consisting of three coupled Markov random fields (MRF's). These three MRF's model a smooth field for depth/disparity, a line process for depth discontinuity and a binary process for occlusion, respectively. After eliminating the line process and the binary process by introducing two robust functions, we obtain the maximum a posteriori (MAP) estimation in the Markov network by applying a Bayesian belief propagation (BP) algorithm. Furthermore, we extend our basic stereo model to incorporate other visual cues (e.g., image segmentation) that are not modeled in the three MRF's, and again obtain the MAP solution. Experimental results demonstrate that our method outperforms the state-of-art stereo algorithms for most test cases.

Motion texture: a two-level statistical model for character motion synthesis

Abstract: In this paper, we describe a novel technique, called motion texture, for synthesizing complex human-figure motion (e.g., dancing) that is statistically similar to the original motion captured data. We define motion texture as a set of motion textons and their distribution, which characterize the stochastic and dynamic nature of the captured motion. Specifically, a motion texton is modeled by a linear dynamic system (LDS) while the texton distribution is represented by a transition matrix indicating how likely each texton is switched to another. We have designed a maximum likelihood algorithm to learn the motion textons and their relationship from the captured dance motion. The learnt motion texture can then be used to generate new animations automatically and/or edit animation sequences interactively. Most interestingly, motion texture can be manipulated at different levels, either by changing the fine details of a specific motion at the texton level or by designing a new choreography at the distribution level. Our approach is demonstrated by many synthesized sequences of visually compelling dance motion.

Advances in strength theories for materials under complex stress state in the 20th Century

Abstract: It is 100 years since the well-know Mohr-Coulomb strength theory was established in 1900. A considerable amount of theoretical and experimental research on strength theory of materials under complex stress state was done in the 20th Century. This review article presents a survey of the advances in strength theory (yield criteria, failure criterion, etc) of materials (including matellic materials, rock, soil, concrete, ice, iron, polymers, energetic material, etc) under complex stress, discusses the relationship among various criteria, and gives a method of choosing a reasonable failure criterion for applications in research and engineering. Three series of strength theories, the unified yield criterion, the unified strength theory, and others are summarized. This review article contains 1163 references regarding the strength theories. This review also includes a biref discussion of the computational implementation of the strength theories and multi-axial fatigue.

Effects of quarantine in six endemic models for infectious diseases.

Abstract: Thresholds, equilibria, and their stability are found for SIQS and SIQR epidemiology models with three forms of the incidence. For most of these models, the endemic equilibrium is asymptotically stable, but for the SIQR model with the quarantine-adjusted incidence, the endemic equilibrium is an unstable spiral for some parameter values and periodic solutions arise by Hopf bifurcation. The Hopf bifurcation surface and stable periodic solutions are found numerically.

Cryptanalysis of a chaotic image encryption method

Abstract: The security of digital images attracts much attention recently, and many image encryption methods have been proposed. In IS-CAS2000, a new chaotic key-based algorithm (CKBA) for image encryption was proposed. This paper points out CKBA is very weak to the chosen/known-plaintext attack with only one plain-image, and its security to brute-force ciphertext-only attack is overestimated by the authors. That is to say, CKBA is not secure at all from cryptographic viewpoint. Some experiments are made to show the feasibility of the chosen/known-plaintext attack. We also discuss some remedies to the original scheme and their performance, and we find none of them can essentially improve the security of CKBA.

Relationships between the microstructure and properties of thermally sprayed deposits

Abstract: Thermally sprayed deposits have layered structure composed of individual splats. The individual splats have quenching microstructure of quasi-stable preferred fine grains. However, this fine-grained microstructure of the deposits is usually not reflected by improved performance of the deposits because a layered structure with two-dimensional voids occurs between lamellar interfaces. The microstructure of the thermal spray deposits with the emphasis on the layer structural parameters is reviewed. Conventionally, one of the most common quantitative parameters used to characterize the microstructure of the thermally sprayed deposits is the porosity, measured by different methods. However, it is illustrated that the relationships between properties and porosity for bulk porous materials processed by conventional processes cannot be applied to thermally sprayed deposits owing to the two-dimensional characteristics of voids. The total porosity in the deposits is not meaningful from the viewpoint of prediction of the deposit properties. An idealized structural model and related parameters, instead of porosity, are proposed to characterize quantitatively the microstructure of the thermally sprayed deposit. The relationships between the properties and the structural parameters are presented for the plasma-sprayed ceramic deposits based on the proposed microstructure model. The properties include the Young’s modulus, fracture toughness, erosion resistance, and thermal conductivity of the plasma sprayed ceramic deposits. The correlations of theoretical relationships with reported experimental data are discussed. An agreement of theoretical with observed values suggests that the lamellar structure of the deposit with limited interface bonding is the dominant factor controlling the performance of the deposit.

The algorithm on knowledge reduction in incomplete information systems

Abstract: Rough set theory is emerging as a powerful tool for reasoning about data, knowledge reduction is one of the important topics in the research on rough set theory. It has been proven that finding the minimal reduct of an information system is a NP-hard problem, so is finding the minimal reduct of an incomplete information system. Main reason of causing NP-hard is combination problem of attributes. In this paper, knowledge reduction is defined from the view of information, a heuristic algorithm based on rough entropy for knowledge reduction is proposed in incomplete information systems, the time complexity of this algorithm is O(|A|2|U|). An illustrative example is provided that shows the application potential of the algorithm.

Example-based caricature generation with exaggeration

Abstract: In this paper, we present a system that automatically generates caricatures from input face images. From example caricatures drawn by an artist, our caricature system learns how an artist draws caricatures. In our approach, we decouple the process of caricature generation into two parts, i.e., shape exaggeration and texture style transferring. The exaggeration of a caricature is accomplished by a prototype-based method that captures the artist's understanding of what are distinctive features of a face and the exaggeration style. Such prototypes are learnt by analyzing the correlation between the image caricature pairs using partial least-squares (PLS). Experimental results demonstrate the effectiveness of our system.

Porous nanocrystalline titania films by plasma electrolytic oxidation

Abstract: Nanocrystalline titania films were prepared by plasma electrolytic oxidation of a titanium alloy at 200–350 V in a Na2CO3 electrolytic solution using a pulsed power supply. XRD, EDS and Field Emission (FE) SEM were employed to characterize the phase, composition, and microstructure of the films. Vickers indentation, nanoindentation and adhesion–tension test were used to evaluate the mechanical properties of the films. The phase, pore size and thickness of the films strongly depend on the applied voltage consistent with the previous reports. The films prepared at 350 V were porous with ∼1 μm pores and the pore walls were composed of 10–20 nm rutile crystallites. The hardness, Young's modulus and bond strength of the film were 0.9±0.2 GPa, 32±4 GPa and 37±3 MPa, respectively. The film exhibited a significant plasticity and ductility compared to the conventional coarse-grained titania ceramics.

Chaotic encryption scheme for real-time digital video

Abstract: In this paper, we propose a novel video encryption scheme based on multiple digital chaotic systems, which is called CVES (Chaotic Video Encryption Scheme). CVES is independent of any video compression algorithms, and can provide high security for real-time digital video with fast encryption speed, and can be simply realized both by hardware and software. Whats more, CVES can be extended to support random retrieval of cipher-video with considerable maximal time-out; the extended CVES is called RRS-CVES (Random-Retrieval-Supported CVES). Essentially speaking, CVES is a universal fast encryption system and can be easily extended to other real-time applications. In CVES, 2^n chaotic maps are used to generate pseudo-random signal to mask the video, and to make pseudo-random permutation of the masked video. Another single chaotic map is employed to initialize and control the above 2^n chaotic maps. Detailed discussions are given to estimate the performance of CVES/RRS-CVES, respectively from the viewpoints of speed, security, realization and experiments.

A linear feedback synchronization theorem for a class of chaotic systems

Abstract: A new synchronization theorem for a class of chaotic systems is presented in this paper. We take the last state variable of drive system as the driving scalar signal. Its linear feedback gain is a function of a free parameter. It is also proved that the global synchronization can be attained through simple linear output error feedback. This approach is illustrated by Chua's circuit and a 4D hyperchaotic system.

Relationship between the expression of iNOS, VEGF, tumor angiogenesis and gastric cancer

Abstract: AIM: To investigate the relationship between the expression of inducible nitric oxide synthase (iNOS), vascular endothelial growth factor (VEGF), the microvascular density (MVD) and the pathological features and clinical staging of gastric cancer. METHODS: Immunohistochemical staining was used for detecting the expression of iNOS and VEGF in 46 resected specimens of gastric carcinoma; the monoclonal antibody against CD34 was used for displaying vascular endothelial cells, and MVD was detected by counting of CD34-positive vascular endothelial cells. RESULTS: Of 46 resected specimens of gastric carcinoma, the rates of expressions of iNOS and VEGF were 58.70% and 76.09%, respectively, and MVD averaged 55.59 ± 19.39. Judged by the standard TNM criteria, the rate of expression of iNOS in stage IV (84.46%) was higher than those in stage I, II, III (Fish exact probabilities test, P = 0.019, 0.023 and 0.033, respectively); the rates of expression of VEGF in stage III, IV (76.0%, 92.31%, respectively) were higher than those in stage I, II (Fish exact probabilities test, P = 0.031, 0.017, 0.022 and 0.019). MVDs in stage III, IV (64.72 ± 14.96, 67.09 ± 18.29, respectively) were higher than those in stage I, II (t = 2.378, 4.015, 2.503 and 2.450, P < 0.05, P < 0.001, P < 0.001, P < 0.05, respectively). In 37 gastric carcinoma specimens with lymph node metastasis, MVD (68.69 ± 18.07) and the rates of expression of iNOS and VEGF (70.27%, 83.78%, respectively) were higher than those in the specimens with absence of metastasis (t = 2.205, χ² = 6.3587, χ² = 6.2584, P < 0.01, P < 0.05, P < 0.05, respectively). MVD and the expressions of iNOS and VEGF were not correlated to the location, size or grade of tumor, nor with the depth of invasion of tumor; MVDs in the positive iNOS and VEGF specimens (59.88 ± 18.02, 58.39 ± 17.73, respectively) were higher than those in the negative iNOS and VEGF specimens (χ² = 6.3587 and 6.1574, P < 0.05, P < 0.05, respectively); thus the expressions of iNOS and VEGF was correlated to MVD, but the expression of iNOS was not correlated to that of VEGF. In addition, of the 46 surviving patients, the 5-year survival rate of patients with positive iNOS or VEGF tumors was significantly less than that of patients with negative iNOS-or VEGF tumors (χ² = 4.3842 and 5.4073, P < 0.05, P < 0.05, respectively). CONCLUSION: The expressions of iNOS and VEGF are closely related to tumor angiogenesis, and are involved in the advancement and the lymph node metastasis; thus MVD and the expressions of iNOS and VEGF may serve indexes for evaluating staging of gastric carcinoma and forecasting its risk of metastasis, which will help establish a comprehensive therapeutical measure of post-operative patients and provide a new approach to tumor therapy.

Unit commitment with identical units successive subproblem solving method based on Lagrangian relaxation

Abstract: When the Lagrangian relaxation based methods are applied to solve power system unit commitment, the identical solutions to the subproblems associated with identical units may cause the dual solution to be far away from the optimal solution and serious solution oscillations. As a result, the quality of the feasible solution obtained may be very unsatisfactory. This issue has been long recognized as an inherent disadvantage of Lagrangian relaxation based methods. In this paper, the homogeneous solution issue is identified and analyzed through a simple example. Based on this analysis, a successive subproblem solving method is developed. The new method combines the concepts of augmented Lagrangian relaxation and surrogate subgradient to produce a good search direction at the high level. The low level subproblems including those corresponding to the identical units are solved successively so that the commitments of the identical units may not be homogeneous in the dual solution. Compared with the standard Lagrangian relaxation method, the new method can obtain better dual solutions and avoid the solution oscillations. Numerical testing shows the new method is efficient and the quality of the feasible solution is greatly improved.

Connections between rough set theory and Dempster-Shafer theory of evidence

Abstract: In rough set theory there exists a pair of approximation operators, the upper and lower approximations, whereas in Dempster-Shafer theory of evidence there exists a dual pair of uncertainty measures, the plausibility and belief functions. It seems that there is some kind of natural connection between the two theories. The purpose of this paper is to establish the relationship between rough set theory and Dempster-Shafer theory of evidence. Various generalizations of the Dempster-Shafer belief structure and their induced uncertainty measures, the plausibility and belief functions, are first reviewed and examined. Generalizations of Pawlak approximation space and their induced approximation operators, the upper and lower approximations, are then summarized. Concepts of random rough sets, which include the mechanisms of numeric and non-numeric aspects of uncertain knowledge, are then proposed. Notions of the Dempster-Shafer theory of evidence within the framework of rough set theory are subsequently formed a...

Purchase allocation and demand bidding in electric power markets

Abstract: The purchase allocation problem is one of the most important problems faced by an electric energy service provider under new market environments. The optimal purchase allocation problem for dual electric power markets and demand bid generation are discussed in this paper. The price volatility is explicitly considered in purchase allocation problems and the sequential nature is modeled by conditional stochastic characteristics. An analytical solution for the optimal allocation is derived with given demand and statistical characteristics of the market prices. The method for generating demand bids for purchaser based on market allocation and price forecasting is then developed. The numerical simulations for market allocation and bid generation are demonstrated based on the actual data of the U.S. California power market.

On the security of an image encryption method

Abstract: The security of digital images has attracted much attention, and many different image encryption methods have been proposed. Yen and Guo (see Proc. IEEE Workshop Signal Processing Systems, p.430-37, 1999) proposed a novel image encryption algorithm called BRIE (bit recirculation image encryption). This paper points out that BRIE is not secure enough from strict cryptographic viewpoint. It has been found that some defects exist in BRIE, and a known/chosen-plaintext attack can break BRIE with only one known/chosen plain-image. Experiments are made to verify the defects of BRIE and the feasibility of the attack.

A fast refinement for adaptive Gaussian chirplet decomposition

Abstract: The chirp function is one of the most fundamental functions in nature. Many natural events, for example, most signals encountered in seismology and the signals in radar systems, can be modeled as the superposition of short-lived chirp functions. Hence, the chirp-based signal representation, such as the Gaussian chirplet decomposition, has been an active research area in the field of signal processing. A main challenge of the Gaussian chirplet decomposition is that Gaussian chirplets do not form an orthogonal basis. A promising solution is to employ adaptive type signal decomposition schemes, such as the matching pursuit. The general underlying theory of the matching pursuit method has been well accepted, but the numerical implementation, in terms of computational speed and accuracy, of the adaptive Gaussian chirplet decomposition remains an open research topic. We present a fast refinement algorithm to search for optimal Gaussian chirplets. With a coarse dictionary, the resulting adaptive Gaussian chirplet decomposition is not only fast but is also more accurate than other known adaptive schemes. The effectiveness of the algorithm introduced is demonstrated by numerical simulations.

Estrogen reduces CCL4- induced liver fibrosis in rats

Abstract: AIM: Chronic liver diseases, such as fibrosis or cirrhosis, are more common in men than in women. This gender difference may be related to the effects of sex hormones on the liver. The aim of the present work was to investigate the effects of estrogen on CCL4-induced fibrosis of the liver in rats. METHODS: Liver fibrosis was induced in male, female and ovariectomized rats by CCL4 administration. All the groups were treated with estradiol (1 mg/kg) twice weekly. And tamoxifen was given to male fibrosis model. At the end of 8 wk, all the rats were killed to study serum indicators and the livers. RESULTS: Estradiol treatment reduced aspartate aminotransferase (AST), alanine aminotransferase (ALT), hyaluronic acid (HA) and type IV collagen (CIV) in sera, suppressed hepatic collagen content, decreased the areas of hepatic stellate cells (HSC) positive for α-smooth muscle actin (α-SMA), and lowered the synthesis of hepatic type I collagen significantly in both sexes and ovariectomy fibrotic rats induced by CCL4 administration. Whereas, tamoxifen had the opposite effect. The fibrotic response of the female liver to CCL4 treatment was significantly weaker than that of male liver. CONCLUSION: Estradiol reduces CCL4-induced hepatic fibrosis in rats. The antifibrogenic role of estrogen in the liver may be one reason for the sex associated differences in the progression from hepatic fibrosis to cirrhosis.