Light field is a representation of full four-dimensional radiance of all rays with spatial and angular information in free space, and capture of light field data enables many new development potentials for computational imaging. The historical development of light field photography is summarized, and typical light field photography devices are categorized in view of capture methods for 4D light field. Based on the principles of light field camera, computational imaging theorem, refocusing theory, synthetic aperture refocusing algorithm, and light field microscopic technology are emphatically described. Finally, the promising application perspectives and existing critical issues of light field imaging are discussed.

We use the methods of harmonic sequences and moving frames to study the harmonic two-spheres in the complex Grassmann manifold G(2,5). Through the construction of holomorphic differential forms on S², we can simplify the moving frames along a harmonic two-sphere in G(2,5). Finally, we give some upper bounds of the Gauss curvature of minimal two-spheres in G(2,5).

We investigate the following quasilinear elliptic equation:

where Ω is a bounded domain in R^N with smooth boundary, Δ_pu=div(| ▽ u |^p-2 ▽ u), N≥3, 2≤p < N, 0<q<1, and p^*= (Np)/(N-p). By using variational methods, we obtain a lower bound of the extremal value λ^*(Ω,p,q) for equation (1_λ), which can be explicitly calculated.

Ammonium polyphosphate and pentaerythritol were used to prepare halogen-free flame retardant BA-VAc copolymer emulsion. The flammability and thermal degradation properties of the copolymer were studied by using limited oxygen index (LOI), vertical burning test (UL-94), and thermo-gravimetric analysis(TG). Morphology of residues was detected by means of scanning electron microscopy. The copolymer has the best LOI value of 28.9 and passes UL-94 V- 0. TG data shows higher and more stable char residue in high temperature range.

We present a concurrent execution method for contral of large quantity displacement actuators for extremely large telescope based on FPGA technology, and the key is to build a control module which generates the control signals of displacement actuators. The construction and simulation using low-level verilog hardware language is described in detail, and the implementation of hardware and software systems is also described.

In order to explore the competitiveness of high-tech industries in Beijing, dynamic shift-share analysis (DSSA) was applied in this study. We conclude that electron and communication equipment manufacturing industries possess a better development condition and a strong competitiveness; medical equipment manufacturing industry and computer and office equipment manufacturing industry lack competition and development potential; and although aerospace and aviation industry and medical manufacturing industry possess certain competitiveness, their economic growths are at disadvantage. In future development, more initiatives should be taken to improve competitiveness of potential high-tech industries in Beijing.

We have analyzed the relationship between the snow spectrum and snow contaminants. It is found that from 350nm to 850nm snow reflectance dramatically decreases with increase in concentrations of contaminants in snow. The results show that snow reflectance at 384nm logarithmically decreases with increase in the snow contaminants while the reflectance at 1495nm exponentially increases with the snow contaminants. Besides, we have selected six bands according to the center wavelength of TM to analyze the relation between snow reflectance and snow contaminants, and we come to the same conclusion as from the hyperspectral analysis. Therefore, this method can be used in combination with TM image to forecast the contaminants in snow.

Using the data of population, GDP, and energy intensity in Jilin province from 1978 to 2009 as input data, we use BP neural network to predict carbon dioxide emissions in 2020 in Jilin province under reference and emission mitigation scenarios. The results show that, if we use carbon dioxide emissions per unit of GDP in 2005 in Jilin province as a reference, CO₂ emissions per GDP in 2020 will be reduced by 55.17% and 58.79% under the two scenarios, respectively. The results also show that, if we use the national average level in 2005 as a reference, CO₂ emissions per unit of GDP in 2020 in Jilin province will be reduced by 35.40% and 40.62%, respectively.

The variable coefficient radial heat conduction problem is studied in the present work. A reasonable computing method for heat conductance is proposed in terms of Fourier law,and the interpolation formulae for interface radius and radial heat conductivity in different coordinate systems are thus derived on the basis of the compatible principle of flux.The new interpolation schemes are evaluated in solution of several typical nonlinear and variable coefficient conduction problems, and the solutions are compared with those obtained using the schemes in the literature.Numerical tests show that the solutions obtained using the new schemes are more accurate,because the comprehensive influences of variable cross-section and variable physical property are fully considered in the new schemes.

We propose a self-adaptive algorithm with convergence detection to reduce the computational complexity of HBP. In the conventional HBP, the computational complexity linearly increases with specified iteration upper bound. We introduce convergence detection to stop the iteration of messages which have already converged to optimal values. Experimental results show that the self-adaptive algorithm reduces computational time by 38% or more, and the computational time is insensitive to iteration upper bound. The convergence detection methodology can be used in other HBP-related applications.

We present systematic analysis and planning of multi-beam antenna based on the features of LEO communication satellite. First, the reasons for adopting multi-beam antenna system are analyzed. Calculations of system parameters within the satellite coverage range are performed based on 7-beam equal-coverage-area scheme. System parameters, like beam width and beam angle, are given. Then, the scheme of array antenna is elaborated and analyzed. Some beamforming and multiplexing schemes are designed and analyzed.

The advanced access content system is a standard for content distribution and digital rights management. It revokes compromised devices by revoking their corresponding device keys and manages them by using the subset cover tree method, which needs many device keys for storing each device. On the reasonable assumption that devices are revoked randomly and uniformly, a modified method is proposed for storage and revoking of the keys. Then, the number of the keys decreases greatly, while the number of the subset covers for legal devices only increases slightly.

Considering that TPM access control mechanism can not be directly applied in virtualization computing, we build the security model for virtual domains based on the dependent relationships of TPM objects and the security requirements of the virtual domains. We add the security constraints of virtual domain, integrity and confidentiality, for the TPM objects' access requests in the model and solve the problems about TPM objects creation, usage, and destroy in multiple virtual domains. The logic analysis for the security rules in the model are further given in this paper. Through the tests on the prototype system, we show that the model has very small performance impact on trust virtualization platform.

The existing covert communication method based on TCP/IP or HTTP often utilizes characteristics of protocol and hides extra data in specific fields of protocol header. Such a method leaves some obvious signatures. However, timing-based covert communications have some kind of traffic pattern. We propose a censorship-resistant covert communication protocol based on Web communication behavior, which combines dynamic asymmetric communication and Markov model-based Web usage prediction. In this paper we focus on covert communication protocol, prototype system, and security of the protocol. The test results show that our method has high performance and safety.

Recent popularity of Web 2.0 application has given rise to a large number of Web vulnerabilities, and XSS vulnerability is among the top security threats. In recent years, the occurrence of XSS worms worsened the situation of Web security. Existing XSS defense methods mainly depend on filtering users’ inputs on the server side, which cannot protect in time the main victims of XSS attacks, the Internet users. In this paper we focus on the analysis of XSS behavior, especially the propagation behavior of XSS worms, and propose a new client-side XSS defense method, StopXSS. The testing experiments show that our method can defend against XSS attacks effectively and can be used to detect even 0-Day XSS worms.

Backtrack4 is a highly evaluated penetration test platform. It contains large database of security tool collection up-to-date, but it can not work efficiently without data supporting. We propose a penetration test framework (PTF) with unified data format mechanism, which can accomplish penetration testing automatically and efficiently. Tools are used automatically for information detection, vulnerability assessment, and report createment. Real network experiments show that PTF can highly enhance the effectiveness of penetration test using Backtrack4.

We propose a DCT domain watermarking algorithm, which is consistent with Kerckhoff principle.This algorithm transforms the watermark image before the watermark embedding and embeds watermark images by key vector, which improves the robustness and security greatly. The experiments indicate that the program has a better robustness on Gaussian low-pass filtering, white noise, JPEG compression, image shearing, and other kinds of attack and a better security.

We propose an algorithm for steganalysis of images based on spread-spectrum data hiding. We employ Markov chains and a combination of threshold method and improved pattern recognition method, and we can discriminate whether an image contains hiding data through detecting the relativity of correlated pixels of the image. Results of the experiments on Corel’s image library show that this algorithm reduces the false alarm rate and increases the precision rate.

Byzantine attacks highly threaten the normal communication of the network. An ontology-based model of Byzantine attacks is proposed based on the detailed analysis of the characteristics of Byzantine attacks. The model is described in the three aspects, class, relation, and reasoning. The model realizes the universal expression of the concepts and achieves information sharing among the various nodes to make them work together. Finally we give an example of the Byzantine overlay network wormholes attack to show the availability of the model.