International Journal of Intelligent Engineering and Systems

7

Authors: Yonghong Yue

This paper presents an approach to evaluate traffic-flow parameters under urban road environment. Virtual line based time-spatial images are used for vehicles counting. Three error impacts in complex urban road environments are discussed and related methods are proposed to improve detection rate. These impacts include traffic congestion impact, environments impact and night vehicle headlight impact. The experimental results on real road videos show that the time-spatial method is robust in complex lighting and traffic environment.

8

Authors: Liu Yaofeng, Wang Zhiliang, Wang Wei, Jiang Xiao

This study discusses the development of an intelligent robot, its emotion decision model and its behavior expression model. According to bionics and facial action code system, six basic facial expressions of the robot are realized using multi-motors cooperative control. On the grounds of linear dynamic system and personality matrixes, emotional model of the robot consists of five elements: reaction dynamic, internal dynamic, emotional dynamic, behavior dynamic and personality. Therefore, the robot uses machine vision, voice recognition and infrared sensors which could make it perceives surroundings. Moreover, it also could detect face, analysis emotion in voice and decide actions intelligently. With emotional interactive model, the robot could interact with human being friendly and harmoniously.

7

Authors: Shasha Yu, Wenwei Wang, Meijuan Li

This paper presents an approach based on radial basis function neural network to extract multi-spectral components from the true-color image. According to the color composition principle of chromatics, the multi-spectral components can be approximated from the three primitive components (RGB) of the true-color image. The radial basis function neural network is used as the multivariate interpolator. To train the kernel function centers of the radial basis function neural network, a method to combine the nearest cluster algorithm and the dynamic mean cluster algorithm is implemented. Experiments have been carried out and the results show that the virtual multi-spectral images created by the proposed network are consistent with the real multi-spectral images acquired by hardware equipments.

8

Authors: Hong Zheng, Xuemin Hu

This paper presents a novel vehicle detection approach based on artificial immune principle. The approach is to build an artificial immune system for vehicle detection from high resolution satellite imagery, where vehicles mimic dangerous antigens and vehicle detection templates mimic antibodies. The artificial immune system uses the antibody network concept inspired from the immune system to learn a set of templates called antibodies for vehicle detection. The details of proposed approaches are described in this paper, and experiments on real satellite images are also given to show the feasibility and performance of proposed method.

9

Authors: Kohji Ohtsuka

A repository of Mathematical Knowledge (MaKR) is designed as an individual tool, which also works as an assistant that helps thinking about the mathematical subjects. MaKR is a mathematical Web-based system whose contents are glued by an original LaTeX-like formatting language (low-)OFM that are designed under the concept of pushdown automaton, whose translator form OFM to HTML is written in Perl and named ofm2html. For the management of resources, a special-purpose Web browser called MaKRV is designed. Main topic in this paper is formatting language (low-)OFM.

7

Authors: Łukasz Saganowski , Michał Chora's , Rafał Renk , Witold Hołubowicz

In this paper we present our original methodology, in which Matching Pursuit is used for networks anomaly and intrusion detection. The architecture of anomaly-based IDS based on signal processing is presented. We propose to use mean projection of the reconstructed network signal to determine if the examined trace is normal or attacked. Experimental results confirm the efficiency of our method in worm detection scenario. The practical usability of the proposed approach in the intrusion detection tolerance system (IDTS) in the INTERSECTION project is presented.

8

Authors: Ting Yuan, Jianqing Ma, Yiping Zhong, Shiyong Zhang

The most common solution for implementing access control into a dynamic group is to use a group key unknown to all but the users in the group. The group key (i.e., session key) is updated for every session throughout the lifetime of the group with the procedure of group key distribution. In this paper, we propose a group key distribution scheme with self-healing property that enables users in a dynamic group to establish session keys over an unreliable network with constrained bandwidth resources. To achieve this, our scheme also has the limited group membership property that there exists an upper bound on the number of users in the group. Through modeling and analysis, we show that our scheme has a better tradeoff between storage and communication overhead as compared to previous work. In addition, we propose a variant of the scheme which enables key recovery from a single broadcast message.

6

Authors: Xiangye Zeng , Jianfei Liu , Jian Wang

Since polarization mode dispersion (PMD) is a major impairment for high bit rate fiber optics systems, it is an important thing for system performance that an proper designed PMD compensate component. In orderto use degree of polarization (DOP) as the feedbacd control signal in PMD dynamic compensation, the relationship beween DOP and differential group delay (DGD) must be ascertain first. Firstly, with a mono-chromatic assumption, the mathematics expression ofDOP is derived in this paper. Then, the relationship beween DOP and DGDm pulse width, and power splitting ratio is analyzde when optic pulse is Gaussian. At last, an experiment is presented, in which 10Gbit/s RZ pseudo random sequence is adopted, and the experimental results prove the validity of theoretical analysis.

8

Authors: Dongfang Chen, Bing Zhou

This paper presents an oblivious watermarking scheme based on contourlet transform and quantization index modulus modulation (QIMM).The scheme has the following features. First, the proposed scheme makes better use of multidirectional characteristic of contourlet transform and embed the watermark into the 4 directional high pass sub bands. Second, the scheme avoids the problem of selection of wavelet filter banks since approximately the same results are obtained with different contourlet filter banks in watermark embedding and extracting stage. Third, unlike the conventional watermarking schemes, watermark error correction is integrated to enhance the quality and normalized cross correlation (NC) value of the extracted watermark. Fourth, through the introduction of the error correction, the proposed scheme tries to view NC as a standard to judge the watermarking algorithm. As a comparison, experiments are conducted to demonstrate that our scheme is more effective than the former one based on lifting scheme wavelet transform.

8

Authors: Xin Xu, Jinguang Gu, Hongbing Zhu, Yimin Qiu

This paper discussed several classical algorithms for vector data compression firstly, and put forward an enhanced vector data compressing algorithm, basing on the analyses of the problems caused by applying these algorithms to the vector map. This algorithm integrates the merits of the classical algorithms, thus can tackle the problem of classical compressing algorithms, such as the distortion of the overall direction in James Algorithm, and the terrain distortion in Douglas - Peucker Algorithm and Light Column Algorithm. An evaluation was developed to test the reliability of this algorithm, and MapInfo data was adopted to show the vector map with the precision of 1:10000, from which the 3rd trunk road of Qingshan District in Wuhan was selected for experiment. Result shows that the proposed algorithm has the least difference in the comparison of the total length and average coordinate of the curve, with corresponding data before compression. And, it has the most similarity in its exterior compared to other classical compressing algorithms. Thus this algorithm could be effective in compressing the vector map. And the experimental research results will hopefully serve as useful feedback information for improvements for vector map compressing in personal navigation system.

8

Authors: Hongmei Kai, Hongbing Zhu, Xiaoli Lin, Kei Eguchi

Neural networks have good learning and associative memory abilities and have been widely applied to various fields. We employed the Backpropagation Neural Network (BPNN) to replace the fuzzy methods of the Intelligent Intrusion Detection, Decision, Response System (I2D2RS) [5] to decide the intrusion. Through this improvement the processing of the system was simplified and the performance of the system was enhanced in the intrusion decision. The effectivities of these improvements were confirmed with the experiments.

9

Authors: Li Li, Jian Huang, Xiaofei Huang, Long Chen

This paper presents a novel approach to detect traffic congestion on roads in a natural open world scene observed from TV cameras placed on poles or buildings. In this system, three algorithms are proposed to estimate the road status. They are based on background subtraction, frame differencing and time-spatial image respectively. The experimental results on real road traffic congestion estimation show that the time-spatial method is the most robust in complex lighting and traffic environment. The detailed algorithms and the comparison results are given in the paper.