Advanced Materials Research Vols. 532-533

Abstract: Distributed Denial-of-Service (DDoS) attacks present a very serious threat to the stability of the Internet. In this paper, an anomaly detection method for DDoS attacks via Behavior Profiles Deviation Degree (BPDD) is proposed. First, the behavior profiles of normal traffic and real-time traffic are constructed using Markov Chain respectively, and then BPDD is designed to measure the discrepancy of the two profiles. Furthermore, TCM-KNN (Transductive Confidence Machines for K-Nearest Neighbors) algorithm is applied to identify attacks by classifying the BPDD samples. The experimental results demonstrate that the proposed method can effectively distinguish normal traffic from DDoS attacks, and has higher detection ratio and lower false alarm ratio than traditional methods.

Abstract: The quantization scheme proposed in this paper uses a nonlinear transform to exploit the visual masking such as brightness masking and contrast masking, etc. It’s based on the IaCLA model derived from the intra-band masking. The application of the proposed strategy to JPEG2000 coding framework is presented, and the experimental results show the proposed approach can efficiently protect low amplitude wavelet coefficient and discriminate the simple edge even in the smooth region.

Abstract: Blind CFA interpolation detection，which identifies the demosaicing method used in digital camera by analyzing output images, provides many efficient tools for digital image forensics．In this paper, we proposes an approach of blind CFA interpolation detection based on the entropy of the correlative coefficients．By solving the pixel matrix equation, the CFA interpolation coefficients are calculated and the entropy of the coefficients are obtained, and they are further fed to SVM classifier to identify forgery. The experimental results show a high accuracy on blind CFA interpolation detection．Compared with existing ones，the proposed method in this paper indicates a better performance on the robnstness especially against lossy JPEG compression.

Abstract: In this paper, a novel method is proposed for solving an open problem of shape matching in content-based image retrieval. In order to prepare for matching two images, our novel method uses an improved segment method to get more accurate contours. And then we find a new way to extract the key contour fragments, so that a shape can be presented as many key contour fragments in this way. It is an effective way to do analysis in image contour. In addition, a similar measure method is proposed to get the correlation rate of two images via subsets of key contour fragments. Finally, experiment results show that our new method can get more robustness than existing methods and achieve a superior matching effect in the standard shape databases MPEG-7.

Abstract: Current total variation method excels at denoising and keeping the characteristics of image edges. However, its ability to retain texture details of smoothing region of image is poor. By combining fractional-order differential theory with total variation method, a new image denoising method is proposed. The new method, while effectively inheriting these advantages, uses the fractional-order differential amplitude-frequency and effectively. Simulation results which we have got show that the new method, on the one hand, can better suppress noise, keep the characteristics of image edges, and retain more texture details than integer-order partial differential methods. On the other hand, the method, above mentioned, is more effective and practical on image denoising than results of PSNR.

Abstract: Applications of data analysis technology in analysis of public transit passenger flow and studying the classification of public transit lines using clustering analysis method are introduced in this paper. Public transit lines in Changsha, as a case, is studied, which if of references for optimizing the operation management of public transit.

Abstract: In order to solve the storage problem of mass graphic data in web, an improved hierarchy of graphic objects is used in this paper. By building up layer object, the system can manage the data for different levels of abstraction. It is different from Object-Oriented Data Base (OODB) which used a relational table as an object to manage data. Uniform operations for the same terrain are offered. Using Object-based Storage Device (OSD) is advantaged at data transmission. The method of storing layer object in OSD is given in the paper. Empirical results show that the mechanism is well-suited to fetch detail data in web.

Abstract: Depending on application requirements, the number of processing nodes in parallel satellite system varies. Currently, fault-tolerant design for satellite system often aims to solve specific problems, lacking in universality. This paper presents a scalable autonomous centralized fault-tolerant (SACFt) architecture. Based on it, functional module framework of each system node and autonomous fault-tolerant strategy are designed. This scheme is insensitive to scale change of the processing nodes. It can ensure autonomous fault tolerance and control node parallelism on demand flexibly at the same time, balancing system reliability, scalability, autonomy, real-time feature and many other factors. The validity and real-time feature of the proposed scheme has been verified and evaluated on the prototype system.

Abstract: Metadata query plays an important role in mass storage system. Efficient indexing algorithm can reduce the time and space which greatly determine the efficiency of mass storage system. Typically, temporal and spatial consuming is immense and volatile in the existing metadata management algorithms. In this paper, a novel metadata indexing algorithm is presented. Metadata query algorithm is based on two-level indexing strategy. The metadata is classified into two categories, that are active metadata and non-active metadata. The Bloom Filter is used to generate binary string for active metadata, and the B-tree is used to establish index of each active partition. While, the suitable hash function is selected for each non-active metadata partition. The results show that the multi-level metadata indexing algorithm can reduce the temporal and spatial costs of metadata query.

Abstract: During teaching of architectural design &construction, it is difficult for teacher shows building unit with 3-D characteristics in two-dimensional plane while students are little imagination. With the growth of virtual reality, the content of architectural design &construction become comprehensive and interactive which is understandable for students. The paper discusses the application of virtual reality in course and realizing method.