20 November 2012, Volume 38 Issue 22
-
Select all|
-
Computer Engineering. 2012, 38(22): 5-9. https://doi.org/10.3969/j.issn.1000-3428.2012.22.002
This paper proposes a novel temporal Unequal Error Protection(UEP) scheme for coding and transmission of real-time video over erroneous channels. The scheme introduces a low complexity frame-level distortion model to estimate video transmission distortion. Then combed with rate control algorithms, it calculates the optimal source coding and channel protection strategy by simply searching of QP frame by frame. Extensive simulations show that compared with previous schemes, the proposed scheme not only has low delay and moderate computation complexity, but also improves the quality of reconstructed video sequences significantly and reduces the fluctuation of quality between different reconstructed frames.
-
Computer Engineering. 2012, 38(22): 10-14. https://doi.org/10.3969/j.issn.1000-3428.2012.22.003
In order to solve the problem of IMS heterogeneous wireless networks handoff delay, an integration resolution of IEEE 802.21-defined Media Independent Handover(MIH) and IMS architechture is proposed. An algorithm of Markov Decision Process(MDP) based multipath redundant transmission handover is presented, through which seamless handover in IMS is realized. NS-2 simulation results prove that the algorithm compared with the traditional vertical handover algorithm, the effective data transmission rate, handover times and handover stability show a clear advantage.
-
Computer Engineering. 2012, 38(22): 15-18. https://doi.org/10.3969/j.issn.1000-3428.2012.22.004
To solve the problem of deterioration of system performance for the three analog front-end non-ideal factors, such as Carrier Frequency Offset(CFO), IQ imbalance and Sampling Frequency Offset(SFO), in Orthogonal Frequency Division Multiplexing(OFDM) system, this paper proposes a novel joint estimation and compensation algorithm, which estimates and compensates CFO and IQ imbalance in time domain, while Residual Carrier Frequency Offset(RCFO) and SFO in frequency domain. Result shows that, compared with some of the traditional algorithms, the algorithm uses fewer data supporting resources and improves the estimation accuracy, even with large values of non-ideal factors, the proposed algorithm is still able to provide a low bit error rate.
-
Computer Engineering. 2012, 38(22): 19-23. https://doi.org/10.3969/j.issn.1000-3428.2012.22.005
In this paper, an effective classifiable prediction mechanism according to the degree of motion vector prediction is suggested for more fast motion estimation, which classifies the motion vectors as no predictable motion vector, medial predictable motion vector or high predictable motion vector, then different proposed search schemes are adopted for motion estimation. Experimental results show the proposed algorithm reduces the prediction error and has a significant computational speedup compared with other algorithms, but offers a similar, even better performance.
-
Computer Engineering. 2012, 38(22): 24-27. https://doi.org/10.3969/j.issn.1000-3428.2012.22.006
Input-dependent loops can cause the problem of path explosion in symbolic execution, a method based on induction variable for analysis of loop dependence is proposed in this paper. By recognizing the loop induction variables and symbolic expressions, it is combined with the bound conditions to generate feasible path constraint for induction variable branches. Induction variable dependence of nested-loop is solved by symbolic affine analysis. It can generate test cases which can cover the branches of induction variables without loop unrolling. Test results show that by this method, two array access violations of the open source tools Libxml2 are found in while loops.
-
Computer Engineering. 2012, 38(22): 28-33. https://doi.org/10.3969/j.issn.1000-3428.2012.22.007
This paper proposes a color palmprint feature extraction algorithm, combined with four classical algorithms and Octonion BP neural network, to extract palmprint line from color palmprint image. The palmprint line image is decomposed by the two-dimensional wavelet, which is used the octonion vector product representation algorithm to match. Experimental results show that palmprint lines can be extracted well by the extraction algorithm, and the recognition rate of the algorithm is 96%
-
Computer Engineering. 2012, 38(22): 34-38. https://doi.org/10.3969/j.issn.1000-3428.2012.22.008As there are several problems for existing Web social relation evaluation methods whose accurate is not high and have poor stability, this paper proposes a Web social relation evaluation method based on both Web page co-occurrence and intra-page sentence co-occurrence. The method assigns weights to the relation between two persons by information about their names co-occurrence in Web pages and co-occurrence in sentences of Web pages. And two weighting functions are designed based on the proposed method. Experimental results show that the method can assign more appropriate weights to relations in Web social networks and the stability of the method is better, and statement co-occur is more important decision function than web co-occur when metric relations weight.
-
Computer Engineering. 2012, 38(22): 39-42. https://doi.org/10.3969/j.issn.1000-3428.2012.22.009According to that recently the research of user interests mostly focus on dividing user into groups according to similar interests. There are still lacking of considering user distribution interests pattern and lacking of methods to measure user distribution interests pattern. Aiming at the problem, this paper proposes a user interest distribution pattern measurement method. It uses Vector Space Model(VSM) to complete user model, quotes the Gini coefficient and Lurenz curve to measure user distribution interest pattern. Result on Movielens dataset verifies the effectiveness of this method.
-
Computer Engineering. 2012, 38(22): 43-46. https://doi.org/10.3969/j.issn.1000-3428.2012.22.010The current software birthmark system only extracts raw program attributes which leads to unsatisfactory performance. To solve the problem, this paper proposes a k-gram software birthmark selection algorithm. A set is constructed by collecting the variants and same function software of a base software. Mutual information is calculated to measure correlation between opcode-gram and variants software category, the higher mutual information value opcode-grams which is considered more distinction and semantic-preserving are chosen as the final birthmarks. Experimental results show that birthmark selection both improves the credibility and the resilience.
-
Computer Engineering. 2012, 38(22): 47-49. https://doi.org/10.3969/j.issn.1000-3428.2012.22.011Based on semi-structured Peer-to-Peer(P2P) system, this paper designs a P2P metadata management mechanism based on interest aggregation. It uses interest aggregation and hierarchical Bloom filters to locate and organize metadata, and routs the metadata query request to different metadata servers. Experimental results show that this mechanism has good adaptability and scalability, and it can improve metadata query efficiency and access performance.
-
Computer Engineering. 2012, 38(22): 50-54. https://doi.org/10.3969/j.issn.1000-3428.2012.22.012This paper proposes a risk evaluation method of software system. By building a Bayesian network, the method analyses the uncertainty contained in the main activities during requirements analysis phase. It determines the sources of risks related to every requirements-risk in the risk list. Then establish a structure-learning algorithm to evaluate the risk effects in order to get the general result of all risks during requirements analysis phase. Simulation result shows that the methods reflect the risk level during the software requirements analysis phase more accurately, list some of the common main requirements risks in the in risks classification, help to carry out the risk management, and enhance the reliability of the software.
-
Computer Engineering. 2012, 38(22): 55-58. https://doi.org/10.3969/j.issn.1000-3428.2012.22.013In application-specific systems on chip, traffic load between different modules is imbalanced and the delay requirements vary largely, thus bandwidth cannot be fully used in uniform links. This paper presents a non-uniform bandwidth allocation approach, which assigns different width of data wires to each link according to traffic patterns and contention probability. The proposed method employs a heterogeneous architecture to allocate wire resources and optimize throughput. Experimental results demonstrate that under uniform traffic pattern, the optimized heterogeneous network achieves 16% wire savings with similar throughput compared with a regular Network-on-Chip(NoC). Under hotspot traffic pattern, it effectively alleviates local congestion and improves system throughput.
-
Computer Engineering. 2012, 38(22): 59-61. https://doi.org/10.3969/j.issn.1000-3428.2012.22.014Traditional Look-up Table(LUT)-based predistortion technology does not take the statistical properties of power amplifier’s input into consideration. A new LUT-based digital predistortion technique combined with the statistical properties of amplitude of Orthogonal Frequency Division Multiplexing(OFDM) signal is proposed in OFDM systems. Simulation results show that compared with traditional LUT-based predistortion technology, the new scheme can effectively suppress the spread spectrum that is caused by the nonlinearity of power amplifier.
-
Computer Engineering. 2012, 38(22): 62-66. https://doi.org/10.3969/j.issn.1000-3428.2012.22.015For the Co-site Interference(CI) problem of vehicular VHF Frequency-Hopping(FH) radio, a comprehensive and in-depth research is conducted, and an optimization design model is established. The CI main influence factors are analyzed. Through experiments and theoretical research on the important parameters, such as interference power, frequency and Bit Error Rate(BER), the CI analysis model of fixed frequency pattern is established. Based on these researches, through analyzing the constraint conditions of VHF networking, a BER computation method of FH system is proposed, and combined with double multi-objective algorithm, an optimization design model based on co-site radios scale and frequency-distribution is constructed. Example analysis shows that the model is the foundation in decreasing FH radios BER and arranging parameters of radios.
-
Computer Engineering. 2012, 38(22): 67-70. https://doi.org/10.3969/j.issn.1000-3428.2012.22.016To settle the problems existing in multimedia multicast congestion control protocols, a multicast congestion control mechanism based on QPID-AVQ is proposed, termed QPID-MCC. QPID-MCC introduces QPID-AVQ queue management strategy and Explicit Congestion Notification(ECN) technology in the bottleneck router, so that enables the router to mark the new arriving packets in accordance with a certain probability computed by QPID-AVQ algorithm. The receivers calculate the expected TCP-friendly rate based on marking probability, and the sender adjusts the sending rate according to the feedback information from the receiver and the minimum bandwidth requirements of multimedia services. Simulation results show that QPID-MCC has a smooth and stable sending rate, good fairness and fast response to congestion, and can ensure minimum bandwidth requirement of multimedia services.
-
Computer Engineering. 2012, 38(22): 71-75. https://doi.org/10.3969/j.issn.1000-3428.2012.22.017Aiming at reducing energy consumption caused by node address in Wireless Sensor Network(WSN), a dynamic address assignment and updating algorithm is developed. The MAC address is designed to consist of cluster address and inner-cluster address, where the cluster address is obtained from its corresponding address reuse area which is generated through node clustering and cluster merging, and the inner-cluster address is assigned with different lengths. Furthermore, a utility-function-based scheme is proposed for inner-cluster address updating where the utility function is designed to consider both current and past energy consumption status of a node. Simulation results show that the proposed algorithm can efficiently decrease the average address length, avoid address conflictions, and reduce the energy consumption caused by node address effectively.
-
Computer Engineering. 2012, 38(22): 76-79. https://doi.org/10.3969/j.issn.1000-3428.2012.22.018Aiming at the problem of the prevailing higher error rate for the existing mobile sensor network localization, this paper presents a new network localization algorithm based on vector machine technology. This algorithm builds the relationship between the node hop-vector and the location by the vector machine technology and the nodes filter the values which can not accord with demands. The nodes hold the sample point set of their original trajectories, calculate and correct their orientation by Newton interpolation method to upgrade their sample point set, and determines their location coordinates by vector machines to achieve the predicted localization of the nodes. Simulation experimental results show that new algorithm has a lower error rate when compared with other algorithms and still keeps a relatively low error rate especially in the case of less anchor nodes and faster movement speed.
-
Computer Engineering. 2012, 38(22): 80-83. https://doi.org/10.3969/j.issn.1000-3428.2012.22.019A superposed training sequence aided disintegrated channel estimation method is proposed for Orthogonal Frequency Division Multiplexing(OFDM) based multi-node Amplify-and-Forward(AF) cooperative communication over frequency-selective fading channel. This method estimates the cascaded Channel State Information(CSI) and the second channel state information by two block training sequence, and calculates the first channel state information. Simulation results show that the proposed disintegrated estimation can be achieved successfully with greatly less training sequence symbols and less time slot and better accuracy and higher practicality.
-
Computer Engineering. 2012, 38(22): 84-87. https://doi.org/10.3969/j.issn.1000-3428.2012.22.020In order to improve the reliability of package transmission, a protocol of Reliable Transmission based on Grid Model(RTGM) is presented. Instead of laying down nodes’ lives by a lot of energy cost, it makes full uses of the size of networks and the transmission range of nodes dividing the net into several grids for the sake of transmission reliability. Furthermore, RTGM introduces random packet loss model to find a reliable node to be the next hop by using the current nodes’ position and energy messages. Specifically a middle node is found on the line of source to Sink. Not only the nearest but also the most reliable node to Sink is chosen. Theory analysis and simulation experimental result show that, compared with AOMDV(Ad hoc On-demand Multi-path Distance Vector routing), RTGM improves the reliability of multi-hop and reduces the energy cost, which avoids the broadcast storm effectively.
-
Computer Engineering. 2012, 38(22): 88-91. https://doi.org/10.3969/j.issn.1000-3428.2012.22.021Due to the large data forwarding, open and vulnerable in Wireless Mesh Networks(WMN), this paper provides an method of flow anomaly detection based on cross-layer data collecting. It calculates the predicted and measured values of the error window after smoothing the center of mass, and gets the deviation of the error which is estimated value, and thus quickly identifies various attacks. Analysis result shows that the proposed method can decrease memory expense, and has less error.
-
Computer Engineering. 2012, 38(22): 92-94. https://doi.org/10.3969/j.issn.1000-3428.2012.22.022If using Guan Haiming’s method to construct conformal iterative function, the result of iterating big integer times can not be obtained in practical applications. In order to solve the problem, this paper proposes two solutions reference to the applications of the discrete logarithm problem, uses the special structure method to search a simple polynomial, and constructs a feasible and efficient conformal iterative function. The efficiency of the function is proved, and two applications of ElGamal encryption scheme and Schnorr signature scheme are given.
-
Computer Engineering. 2012, 38(22): 95-98. https://doi.org/10.3969/j.issn.1000-3428.2012.22.023In an improved multivariate signature scheme named W-scheme, affine transformation N is used to replaed affine transformation T, and the public key to participate in verifying via replacing T with an affine transformation, but N can be turned to a new affine transformation with T, so the scheme can not hide the signature structure . Aiming at the security loop, by depicting the relationship of the public key and , the structure of can be determined by , so that attack is realized, which proves that adding the secret affine transformation can not lift the level of security of multivariate public key scheme.
-
Computer Engineering. 2012, 38(22): 99-103. https://doi.org/10.3969/j.issn.1000-3428.2012.22.024This paper researches an user authentication scheme of Wireless Sensor Network(WSN) proposed by He et al, points out that it has some security weaknesses such as lacking user anonymity, vulnerability to user impersonation attacks and GW-node bypassing attacks, and provides an improved user authentication scheme based on one-way hash function and symmetric cryptosystem. Analytical results prove that the scheme not only achieves user anonymity, untraceability and entity authentication, but also can resist off-line dictionary attacks, user impersonation attacks and GW-node bypassing attacks. Compared to other related authentication schemes, it satisfies more reasonable computational consumption.
-
Computer Engineering. 2012, 38(22): 104-106. https://doi.org/10.3969/j.issn.1000-3428.2012.22.025Intrusion detection methods based on abnormity need a normal data set to establish the normal behavior mode, but there is not an effective method to get the data set. The number of clustering needs to be initialized in Particle Swarm Optimization(PSO) fuzzy clustering algorithm, and this number is confirmed according to experience, so its accuracy is lower. To solve these problems, this paper proposes a self-control PSO fuzzy clustering algorithm, getting the training sets from the network data. It sets the particle swarm with different number of clustering and the number can be adjusted by control-vector according to the validity function. And the data can be converged into an appropriate number of clustering. Experimental results show that the method can improve the veracity of clustering, and reliable training sets can be got from the network data.
-
Computer Engineering. 2012, 38(22): 107-120. https://doi.org/10.3969/j.issn.1000-3428.2012.22.026In order to improve the security of information transmitted in a multicast network, this paper proposes a network coding scheme based on information security. A large prime number and its primitive root produce different noise symbols, and the symbols are mixed with the original information to achieve the purpose in hiding information. Analysis results show that the scheme is able to make network coding a very high probability of information theory safety by swallowing a small amount of bandwidth. When a dedicated secure channel between the source and sink is usable, the secure channel algorithm can achieve the probability 1 of information theory security without increasing the overhead traffic.
-
Computer Engineering. 2012, 38(22): 111-113. https://doi.org/10.3969/j.issn.1000-3428.2012.22.027Digital watermark may encounter various attacks, such as data compression, feature increasing, random point increasing and map cutting, when it is used in practice. This paper proposes a digital watermarking algorithm based on feature point layer and minimum quadtree division, which is suitable for polyline and polygon layer. Douglas-Peucker compression algorithm is used to extract vector data of feature point layer, and the watermark is embedded into the feature point layer by minimum quadtree division. Experiments show that the algorithm is fairly good in imperceptibility and more robust than MQUAD algorithm in against map cutting attack.
-
Computer Engineering. 2012, 38(22): 114-118. https://doi.org/10.3969/j.issn.1000-3428.2012.22.028In order to make up the insignificance of Web Application Firewall(WAF) with single node in detection efficiency and stability, a WAF based on distributed and P2P architecture is designed and implemented by using reverse proxy technology. Reverse proxy technology is used to response the Web request. Aiming to implementing P2P architecture, the same program is run on each node. Master and auxiliary nodes are determined dynamically by demands. The master node has the characteristics of session keeping and load balance, and auxiliary nodes detect messages by expert library and plug-in components. Experimental results show that the system can effectively prevent attacks from application layer, and it is more efficient and stable than single node.
-
Computer Engineering. 2012, 38(22): 119-121. https://doi.org/10.3969/j.issn.1000-3428.2012.22.029By analyzing the properties of braid group and some decision problems on braid group, this paper proposes a protocol by applying twisted conjugacy search problem, subgroup membership decision problem and root search problem on to specific subgroups of braid groups where the subgroups enjoy unsolvable word problem. Security analysis shows that the protocol can resist length attack, key-only attack, chosen message attack and chosenplaintext attack and so on.
-
Computer Engineering. 2012, 38(22): 122-125. https://doi.org/10.3969/j.issn.1000-3428.2012.22.030The participants’ weight in the previous secret sharing scheme is fixed, which can not meet the practical application. Therefore, a new secret sharing scheme that the participants’ weight can dynamically change is proposed based on the trust evaluation model. Making adjustments to the secret shadow number and changing the secret sharing function in accordance with different behaviors of participants. After removing the dishonest member, the new members can be admitted. Under conditions of not changing participants’ shared secret, participants’ secret shares are prior to be updated. Analysis result shows that the scheme is a secure secret sharing scheme in which the participants’ weight and secret shadow can dynamically change.
-
Computer Engineering. 2012, 38(22): 126-129. https://doi.org/10.3969/j.issn.1000-3428.2012.22.031The traditional particle filter tracking algorithm usually leads to tracking error or failure, when the target is interfered by the similar background or occlusion by the other object, and fast-moving environment. In order to track moving objects in complex outdoor scene, the particle filter tracking method with histograms of oriented gradients and Autoregressive Moving Average(ARMA) model fusion is proposed for its robustness. Various moving objects have different motion state in different scenarios, so a kind of ARMA motion model is introduced which forecasts the motion state of next frame based on motion states of former two frames, and overcomes the posture change and occlusion interference. Histograms of Oriented Gradients(HOG) observation model which can describe object’s texture characteristic are applied to solve track problem in the complex scene that illumination is uneven or illumination suddenly change leads to mutation of object color.
-
Computer Engineering. 2012, 38(22): 130-132. https://doi.org/10.3969/j.issn.1000-3428.2012.22.032This paper presents a system for ellipsis identification in Chinese which is based on machine learning. The system can be used to select a number of features and feature combinations through preprocessing the corpus. And Chinese ellipsis identification can also be achieved by the ellipsis identification model built by Support Vector Machine(SVM). The performance of the system in different parser tree is studied as well. Experimental result shows that the system has F value of 84.01% on the standard parser tree, and 68.22% on automatic sentence parser tree.
-
Computer Engineering. 2012, 38(22): 133-136. https://doi.org/10.3969/j.issn.1000-3428.2012.22.033This paper analyzes the two minimum time planning algorithms. In allusion to deficiencies in the application of the algorithm, based on the SPFA static planning algorithm, combined with the advantages of the two planning algorithms, it proposes a minimum time path planning improved algorithm. Example results show that the algorithm can provide path planning with real-time, high efficiency and predictability. It better meets the needs of users in urban traffic.
-
Computer Engineering. 2012, 38(22): 137-140. https://doi.org/10.3969/j.issn.1000-3428.2012.22.034Aiming at the lane problem in the cross intersection of two-dimensional traffic network with traffic lights, this paper puts forward a four-direction traffic model, and it includes the impact of acceleration parameters and change lanes probability on traffic flow, takes advantage of the methods of cellular automaton of BML model so as to simulate the whole dynamic process of vehicle, and points out the relationship between the critical point of lane changing probability and traffic flow influence. Numerical results show that the probability of lane change has disturbance impact on input and output traffic flow, and lane changing rules are correspond to the basic line of vehicles running.
-
Computer Engineering. 2012, 38(22): 141-145. https://doi.org/10.3969/j.issn.1000-3428.2012.22.035In order to improve the real-time performance of Case-based Reasoning(CBR) reasoning mechanism, a new reasoning mechanism of intelligent robot is proposed in this paper. This paper gives the CBR reasoning process, uses the time-limited algorithm and space vector algorithm to improve it, so to speed up the search and complete the study, and uses modular parts to build intelligent platform. Experimental results show that this mechanism can use less time in perform the same task, and get more reasonable result than CBR.
-
Computer Engineering. 2012, 38(22): 146-150. https://doi.org/10.3969/j.issn.1000-3428.2012.22.036According to the low convergence speed, poor accuracy and stability of Particle Swarm Optimization(PSO) algorithm, an improved PSO algorithm based on fitness feedback effect is proposed in this paper. By calculating and normalizing the variances of fitness in the neighboring iterations and then imposing feedback on inertial weights, the fitness oscillation is weakened and the ability of avoiding local-optimization is enhanced during the optimizing process. Test results show that the global search ability is improved, and this algorithm has high convergence speed and stability.
-
Computer Engineering. 2012, 38(22): 151-153. https://doi.org/10.3969/j.issn.1000-3428.2012.22.037In order to improve the robustness of face recognition to the changes of facial pose, position and expression, a facial feature extraction method based on Nonsubsampled Contourtlet Transform(NSCT) and Modified Pulse-coupled Neural Network(M-PCNN) is proposed in this paper. By using NSCT, the input images are decomposed into a number of sub-images with various scales and directional features. The different subbands are decomposed into a sequence of binary images by using M-PCNN. The information entropies of each binary images are calculated and regarded as facial features. A Support Vector Machine(SVM) classifier is employed to implement recognition and classification. Simulation results show that this method has good robustness, and can achieve better result in verification and classification.
-
Computer Engineering. 2012, 38(22): 154-158. https://doi.org/10.3969/j.issn.1000-3428.2012.22.038For the complexity of generation process resulted from the difficulty of standardizing the extraction method, this paper proposes an Event Condition Action(ECA) rule extraction method based on scenario analysis. It defines the template specification of event, condition and action as well as syntactic structure of ECA rules, uses the scenario analysis method to put forward a method of extracting ECA rules from scenarios and detect the consistency. Example results show that this method is to standardize the extraction process and ensures the accuracy of ECA rules.
-
Computer Engineering. 2012, 38(22): 159-162. https://doi.org/10.3969/j.issn.1000-3428.2012.22.039To overcome the defects that the traditional Scale Invariant Feature Transform(SIFT) method is sensitive to parameters applied, a SIFT feature point matching algorithm based on field intensity and convex hull is proposed. Based on the traditional SIFT matching method and the convex hull of the point sets in template and observe images, a concept of the gravitational field intensity in physics is introduced to depict the space relationships between features points for image matching in this paper. The proposed algorithm makes full use of the space geometry relationship between feature points in matching process. Experimental results show that, by using the proposed algorithm, the correct rate of matching between the point sets can be greatly improved, and can get more feature matching point.
-
Computer Engineering. 2012, 38(22): 163-166. https://doi.org/10.3969/j.issn.1000-3428.2012.22.040This paper presents a Chinese Webpage keyword extraction algorithm based on word extension model. It creates an evaluation function to transform term-document matrix by scoring candidate keyword based on its Web structure, part-of-speech, length, TF-IDF value, and uses the word extension model to extend the candidate keywords into key phrases which is based on the n-gram language model. Experimental results show that the proposed algorithm has better performance compared with the traditional keyword extraction algorithms.
-
Computer Engineering. 2012, 38(22): 167-170. https://doi.org/10.3969/j.issn.1000-3428.2012.22.041In Multi-label K-nearest Neighbor(ML-KNN) learning algorithm, the number of nearest neighbors is given in prior and its value is chosen without considering the distribution of samples, it is possible that highly similar samples are not in the nearest neighbor or low similar samples are in the nearest neighbor set, which affect the performance of the classifier. In view of this case, a novel ML-KNN algorithm is put forward based on the idea of Granular Computing(GrC), the nearest neighbor set is constructed with the controlling of the granular hierarchy, and the nearest neighbors of a sample have high similarity and highly similar samples can be added to nearest neighbor set. Experimental results show that most of the evaluation criteria in new algorithm are better than the traditional algorithm.
-
Computer Engineering. 2012, 38(22): 171-175. https://doi.org/10.3969/j.issn.1000-3428.2012.22.042The primary difficulty of text clustering lies in the multi-dimensional sparseness of texts. A short text clustering algorithm which takes semantic and statistic features into account is proposed. A dimensionality reduction is achieved via the semantic relativity analysis of lexical semantics by semantic dictionary. The second dimension reduction is completed after a feature selection through statistical methods. The short text clustering is obtained with the combination of the two reductions. Experimental result shows that the algorithm has better clustering effect and efficiency on short text.
-
Computer Engineering. 2012, 38(22): 176-178. https://doi.org/10.3969/j.issn.1000-3428.2012.22.043A variable illumination face recognition method based on Curvelet and Local Binary Pattern(LBP) is proposed to solve the recognition rate decline problem because of the variant illumination and noise. The original face images are decomposed by Curvelet transform. It processes the coefficients of the first layer by logarithm computing and LBP method to remove the effect of the illumination, and directly discards the surplus highest frequency information to alleviate the environment noise and the fake information at the intersection of the shadow. The face recognition is done by the Principal Component Analysis(PCA) and Linear Discriminant Analysis(LDA) using the preprocessed the first layer data and the other layer data. Simulation results show that this method can alleviate the effect of the illumination and environment noise, and has good robustness.
-
Computer Engineering. 2012, 38(22): 179-182. https://doi.org/10.3969/j.issn.1000-3428.2012.22.044To address the frequency resolution insufficiency of Short-term Fourier Transform(STFT), a novel low-resolution face recognition method is proposed in this paper by combining Dual-tree Complex Wavelet Transform(DTCWT) and compact Local Binary Pattern(LBP) histogram. Face images are processed using DTCWT to obtain multi-scale and multi-direction frequency response, and then face features are conducted by applying LBP histogram. In addition, a statistical uniform pattern is introduced to improve the efficiency of the proposed algorithm. Experimental results show that this method has excellent recognition accuracy on low-resolution face recognition.
-
Computer Engineering. 2012, 38(22): 183-185. https://doi.org/10.3969/j.issn.1000-3428.2012.22.045The float exponential operation of bilateral filtering is not suitable for hardware implementation. In order to solve this problem, a modified bilateral filtering algorithm based on similarity judgment is proposed in this paper. Photometric similarity of every pixel in the filtering window is evaluated. Then these similar neighboring pixels are filtered by the modified bilateral filtering, which adopts reciprocal function instead of exponential function. Experimental results show that the modified bilateral filtering algorithm can balance the details and the flat areas, and it facilitates hardware implementation. The average Peak Signal Noise Ratio(PSNR) is 0.27 dB higher than bilateral filtering algorithm.
-
Computer Engineering. 2012, 38(22): 186-189. https://doi.org/10.3969/j.issn.1000-3428.2012.22.046This paper proposes a novel image registration algorithm based on Zernike moments and multilevel feedforward neural networks. Low order Zernike moments are used as image global pattern features and the algorithm feeds them into multilevel feedforward neural networks to provide translation, rotation and scaling parameters. The second level feedforward neural network is utilized behind the first level neural network to improve parameter estimation accuracy. Experimental results show that the proposed algorithm can provide accurate registration and robust to noise attack.
-
Computer Engineering. 2012, 38(22): 190-193. https://doi.org/10.3969/j.issn.1000-3428.2012.22.047Aiming at the images with neat boundaries and the additive noise satisfies the condition that its distribution is nonGaussian such as uniform, salt & pepper, Laplace and so on, a GNC method for image restoration based on l1-nonconvex nonsmooth function is presented taking the ideas of fitting gradually and variables conversion. According to the original nonconvex nonsmooth function, it establishes a series of fitting functions, which can get better initial value of original problem, meanwhile variable conversion is used effectively to deal with the nondifferen- tiability of l1-nonconvex nonsmooth function. Experimental results show that the method has higher restoration performance.
-
Computer Engineering. 2012, 38(22): 194-197. https://doi.org/10.3969/j.issn.1000-3428.2012.22.048On the base of the classic total variation de-noising model and its improved models, this paper proposes an improved double-entry adaptive Total Variation(TV) model based on local information of pre-made images. An adaptive spatial fidelity term is intended to ease the smoothing effect by the second-order nonlinear filtering over the details. An adaptive regularization term is to further reduce the staircase effect, also to achieve a more stable and convergence value. Experimental results show that the improved method still can achieve better results in high noise. Compared with the original method, it has better noise robustness.
-
Computer Engineering. 2012, 38(22): 198-200. https://doi.org/10.3969/j.issn.1000-3428.2012.22.049To overcome the image distortion by camera in the image level detection system, an image distortion correction algorithm based on image plane is proposed. It uses a virtual standard calibration graph rather than reality one. It uses second-order radial model and iterative striking distortion coefficient, and bilinear interpolation to interpolate the output pixel gray level. Experimental results show that the method can correct distorted image effectively, and the error of distorted correction is within 0.35 pixel. It can improve the accuracy of video detection greatly.
-
Computer Engineering. 2012, 38(22): 201-204. https://doi.org/10.3969/j.issn.1000-3428.2012.22.050The serious effect of multiplicative speckle noise makes the understanding of Synthetic Aperture Radar(SAR) images very hard. A segmentation algorithm for SAR images based on component tree is proposed. This algorithm builds component trees for SAR images. It gives a kind of adaptive non-local criterion based on global feature to merge the similar components in component trees. It keeps the most significant components to filter the images and get the targets. Experimental results show that this algorithm can get good segmentation and maintain detail feature of targets.
-
Computer Engineering. 2012, 38(22): 205-207. https://doi.org/10.3969/j.issn.1000-3428.2012.22.051Image retrieval based on relevance feedback algorithm by the iteration point will fall into the local optimum. So an adaptive relevance feedback algorithm is proposed, which is able to address such a problem mentioned above. If query point reaches optimum, this query point is automatic splitted into two query points using modified EM algorithm. Two points as original query points, and then reach local optimum respectively. If users do not satisfy accuracy, further more these two points are splitted into four query points, and achieve local optimum. The process of iteration is not stopped until a user is satisfied. Experimental results show that, compared with other algorithms, the algorithm has higher precision rate.
-
Computer Engineering. 2012, 38(22): 208-210. https://doi.org/10.3969/j.issn.1000-3428.2012.22.052In processing of the digital camera image, automatic white balance is used for partial color revised under different color temperature. In many cases, ordinary algorithms for auto white balance can not return true color images exactly. Aiming at this problem, this paper introduces an automatic white balance algorithm based on Voronoi figure. It uses Voronoi figure to statistics white point number, estimates immediate environment color temperature value, does gain calculation, and corrects images according to the standard images. Experimental results show that this algorithm can improve the image quality in different environment.
-
Computer Engineering. 2012, 38(22): 211-215. https://doi.org/10.3969/j.issn.1000-3428.2012.22.053The Locally Linear Embedding(LLE)-based high-resolution image reconstruction method has some problems about noise sensitivity and discontinuous boundary between image blocks. Aiming at these problems, this paper proposes a manifold learning method for reconstructing high-resolution image through neighborhood nonnegative linear combination. It restricts linear combination coefficients of neighbors to be nonnegative in the reconstruction process of high resolution images based on manifold learning, and uses a new feature extraction method. Experimental results show that the proposed algorithm can reconstruct more detail features, remove discontinuous boundary and obtain higher Peak Signal Noise Ratio(PSNR).
-
Computer Engineering. 2012, 38(22): 216-219. https://doi.org/10.3969/j.issn.1000-3428.2012.22.054The current skeleton-driven deformation algorithms usually use single bone to drive deformation, and the skeleton default is very complex. In order to solve this problem, an improved mesh deformation algorithm based on skeleton is proposed. It combines the Reeb Graph(RG) skeleton extraction theory to obtain the local area of each skeleton, approximates the skeleton nodes into a Bézier curve. With the interaction of skeleton node, this algorithm can automatically adjust connected skeleton nodes, and implements the intuitive local deformation of mesh model. Experimental results show that this algorithm can obtain more natural and smooth deformation results.
-
Computer Engineering. 2012, 38(22): 220-224. https://doi.org/TN911TN911.73.73The Scalable Vector Graphics(SVG) does not support 3D model. Aiming at this problem, this paper introduces the realization of the principle of SVG expand element , researches how to set up 3D model in SVG which is 2D vector graphics, and uses interpolation algorithm with a triangulation network to solve the problem that element does not support points not consistent, not even object. It establishes 3D Digital Elevation Model(DEM) in SVG, and animates it online in changing view points. Approaches of 3D DEM creation by embedded JS and Java daemon are compared, and experimental results show that the embedded JS way cost more time than Java daemon solution. It is necessary to expand the as a basic SVG element to create 3D models.
-
Computer Engineering. 2012, 38(22): 225-227. https://doi.org/10.3969/j.issn.1000-3428.2012.22.056A novel mismatch removal method is proposed to deal with the problem of epipolar rectification in the semi-calibrated case. It discards the mismatch with global scaling and rotation consistency, and rejects more incorrect matches accoding to the information of scale, orientation and spatial layout. The rest correspondences satisfy epipolar geometry constraint, so the method uses least square to estimate the uncalibrated parameters. Experimental results show this mismatch removal method works well in the problem of semi-calibrate epipolar rectification. The rectification precision is high and the image loss is low.
-
Computer Engineering. 2012, 38(22): 228-231. https://doi.org/10.3969/j.issn.1000-3428.2012.22.057To obtain better multi-objective performances of the Interbay material handling system in wafer fabrication, a novel hybrid intelligence-based dispatching approach for the Interbay material handling system is proposed. The weight coefficients of wafer lots’ transportation cost model are adjusted adaptively by Takagi-Sugeno based fuzzy logic method. The optimized dispatching solution between wafer lots and transport vehicles is obtained based on Hungarian algorithm. Experimental results demonstrate that the proposed approach has better comprehensive performances of the Interbay material handling system and SMS than traditional dispatching approaches.
-
Computer Engineering. 2012, 38(22): 232-235. https://doi.org/10.3969/j.issn.1000-3428.2012.22.058This paper proposes a method to calculate integral images on part of the image instead of global scope by hardware, thus avoiding most time-consuming pretreatment. The cost of doing so and the speed of single rectangular feature calculation are acceptable due to the characteristic of AdaBoost that few rectangular features are needed to drop out most unrelated images. With traditional AdaBoost, this method can accelerate the analysis of a great number of high-resolution images. As result, when irrelative image occupies 98% of whole, the performance of this approach will be 31.8% higher than traditional ones.
-
Computer Engineering. 2012, 38(22): 236-239. https://doi.org/10.3969/j.issn.1000-3428.2012.22.059Aiming at the problem that start-stop scheme of the pumps is uncertain and irregular, which cause excessive power consumption. This paper proposes an optimal operation model whose objective function is the minimum energy consumption of the drainage system, using the hybrid genetic algorithm to find the optimal solution. Simulation results show that this method can effectively avoid the premature of genetic algorithm, and improve the local searching capability, as well as has strong applicability, and the average energy saving efficiency of the drainage system is above 6.5%.
-
Computer Engineering. 2012, 38(22): 240-243. https://doi.org/10.3969/j.issn.1000-3428.2012.22.060This paper designs motion control system of autonomous monitoring ship for lake water environment. The navigation system is designed using GPS and electronic compass, ultrasonic sensors are adopted for avoiding obstacles. The autonomous power supply system using solar power and battery is given. The automatic parking system composed of ultrasound, infrared sensors and mechanical clasp is established. Based on coordinate transformation, the kinematic model of autonomous ship is formulated. The hierarchical controller structure and sliding model control algorithm are designed for the speed control of double motors. Experimental results show that sliding model control can adapt to the mutation interference of flow, and turning is smooth. Autonomous ship can achieve automatic self-shipping berths and energy self-sufficiency. The navigation and control systems are valuable and velocity response time is less than 4 s and large angle rotation response time is less than 13 s.
-
Computer Engineering. 2012, 38(22): 244-247. https://doi.org/10.3969/j.issn.1000-3428.2012.22.061An improved tracking and localization algorithm for mobile stations based on Extended Kalman Filtering(EKF) is proposed. In this algorithm, EKF is used to obtain multiple tracking trajectories of a mobile station with its initial position estimation being obtained. Combining the technique of removing the trajectories with larger deviations and data fusion with weighted averaging, a better trajectory among the trajectories is found. Based on this, a distance threshold is coordinated with the matching management of the better trajectory’s points for smoothing the better trajectory to obtain a best tracking trajectory. Simulation results show that the algorithm has low computational complexity, strong robustness as well as higher localization accuracy compared with traditional EKF tracking and localization algorithms.
-
Computer Engineering. 2012, 38(22): 248-250. https://doi.org/10.3969/j.issn.1000-3428.2012.22.062Earlier serial port and Ethernet interface conversion is realized by pure hardware. It has features of high data transmission speed, but function single as well as poor flexibility. Due to these disadvantages, interface conversion system based on ARM9 is proposed. MOXA UC-7101 is utilized as hardware. Application program is developed under μCLinux embedded OS. System communication follows modbus protocols. When net is connected, SD card data can be retrieved correctly according to net instruction, and data between monitor center and lower RTU can be transferred correctly and timely. System test results show that MOXA begins to poll RTU and response data is stored into SD card while net is shut off.
-
Computer Engineering. 2012, 38(22): 251-259. https://doi.org/10.3969/j.issn.1000-3428.2012.22.063In order to improve the computational efficiency of the Hybrid Soft Tissue Deformable Model(HSTDM), a Precomputation-based Hybrid Soft Tissue Deformable Model(P-HSTDM) is proposed. The proposed model uses HSTDM model to obtain the exact discrete state space data by offline precomputation, then the smoothed particle hydrodynamics interpolation function is used to calculate the force feedback and the offset value of deformation node in real-time interaction. Simulation results show that P-HSTDM model can describe linear, nonlinear and anisotropic material properties of the soft tissue, and provide the refresh rate of visual and force feedback above 60 Hz and 500 Hz, respectively.
-
Computer Engineering. 2012, 38(22): 255-259. https://doi.org/10.3969/j.issn.1000-3428.2012.22.064An algorithm based on coefficient matrix transformation for computing Mixed Polarity Reed-Muller(MPRM) expansions of Boolean functions with multiple outputs is proposed. The optimal MPRM is solved by taking the number of product terms in MPRM as minimization criterion and exercising exhaustive strategy for polarity space exploring. The proposed method is used to solve the optimal MPRM of MCNC and ISCAS benchmarks, and compared with the method using tabular technique, test results show that the proposed coefficient matrix transformation algorithm achieves 55.8% performance improvement on average for optimal MPRM solving.
-
Computer Engineering. 2012, 38(22): 260-262. https://doi.org/10.3969/j.issn.1000-3428.2012.22.065Reconfigurable systems are very efficient on computing intensive domains, and critical loops of algorithms can be set up on a reconfigurable array especially. But how to mapping applications is still a hard work which limits the development of reconfigurable technique. This paper proposes a way to define timing models and method to modify the reconfigurable mapping performance. It is meaningful to the compiler designs. Memory wall on a hardware array and data flow graph topology of algorithms are considered, and the setup/hold times of a loop mapping process is also derived. The definition of path’s timing feature gives a new way to take out a reconfigurable compiler. Verification shows 43% performance improvement is achieved on mapping critical loop deblocking of H.264 decoding to a reconfigurable system.
-
Computer Engineering. 2012, 38(22): 263-266. https://doi.org/10.3969/j.issn.1000-3428.2012.22.066An improved data integration method is proposed, in order to solve the problem of inadequate use of data-related characteristics in the procedure of constructing gene regulatory networks based on data integration. It uses knockout data and perturbation data to make predictions respectively, gives different confidence to the predictions of the two experimental data according to their overlap degree, and priority to the part with higher overlap degree. It sorts the predictions in accordance with their confidence. To assess the performance of the two methods, Dream3(Dialogue for Reverse Engineering Assessments and Methods) datasets are used. Experimental results show that the overallscore of the improved method is 4.9% higher than the existing method.
-
Computer Engineering. 2012, 38(22): 267-270. https://doi.org/10.3969/j.issn.1000-3428.2012.22.067Aiming at the increasingly high performance requirements of text retrieval and computational biology, this paper proposes a bit-parallel approximate string matching based on Compute Unified Device Architecture(CUDA). Taking advantage of highly parallel computing architecture and high memory bandwidth characteristics of Graphic Processing Unit(GPU), it researches and realizes the acceleration of Bit-parallel dynamic Programming Matrix(BPM) algorithm based on CUDA through improving data accessing and computing strategy, and a comparison test about the GPU accelerated performance has also been carried out. Experimental results show that BPM algorithm can achieve 20 times speedup by GPU.
-
Computer Engineering. 2012, 38(22): 271-275. https://doi.org/10.3969/j.issn.1000-3428.2012.22.068In accordance with the problems in automatically generating application based on Internet Communications Engine(ICE), the ICE system modules are presented, including the ICE-based application of independent and distributed systems. This paper designs a programming-based code generator from eXtensible Markup Language(XML) and template technology, which supports nested loops and selection statements, and provides the concept of chain-out template and data field conversion. Experimental result shows that users can quickly generate custom ICE framework described by the system modeling or requirement description, turn their attention to the business logic programming, and improve development efficiency.
-
Computer Engineering. 2012, 38(22): 276-278. https://doi.org/10.3969/j.issn.1000-3428.2012.22.069Aiming at overcoming the traditional Bag of Visual Word(BoVW) model’s sensitivity to image scale’s variation, this paper proposes an image annotation method based on improved BoVW model. It incorporates with multiple spaces information and transfers original images into multiple scale spaces and constructs multiple scale vocabularies. Images are represented as a family of feature histograms with different scale. Multiple kernel learning is introduced to optimize the histograms weights of different scale in order to acquire discriminative classifying power. Experimental results prove the validity of the method, it outperforms BoVW on image annotation precision ranged from 17.8% to 25.7%.
-
Computer Engineering. 2012, 38(22): 279-282. https://doi.org/10.3969/j.issn.1000-3428.2012.22.070The segmentation performance on Chinese analyzer of Lucene is insufficient, and the third party analyzer is difficult to choose. Because of this problem, this paper introduces several kinds of support Lucene analyzer, based on the experiment, sentence segmentation, word segmentation speed, index space and time, retrieval results and speed of retrieval are compared and researched. Analysis results show that, in Lucene framework, Paoding analyzer based on dictionary segmentation has the best overall performance, one-word analyzer of Lucene has the highest segmentation speed, imdict and ICTCLAS4J analyzer have greater room for improvement on the algorithm efficiency.
-
Computer Engineering. 2012, 38(22): 283-286. https://doi.org/10.3969/j.issn.1000-3428.2012.22.071Traditional NSGAⅡalgorithms are easy to get local optimization and less diversity of the population. Aiming at the problems, this paper proposes a balanced optimized delay model and the Boltzmann-NSGAⅡ mapping algorithm. The delay model is constructed from the macroscopic link load distribution and the single node queue latency. The Boltzmann mechanism, multi-point crossover operator, random mutation operator are introduced into the algorithm to maintain the diversity of the population and increase the local search capability. Simulation results prove that the optimal solution set obtained by the Boltzmann-NSGAⅡ algorithm is better in the convergence and uniformity, and the mapping effect of the Boltzmann-NSGAⅡ algorithm is better.
-
Computer Engineering. 2012, 38(22): 287-290. https://doi.org/10.3969/j.issn.1000-3428.2012.22.072An efficient algorithm of constructing Delaunay triangulation network based on discrete point set is presented. A large scale data points are pre-processed by sorting with blocks, which makes the next inserting point closely neighboring the newly inserting one, so the algorithm is in accord with the space related theory. When it comes to predicate the location of the next inserting point in which triangles is very time-consuming, it adopts an efficient and robust blending algorithm of the shortest path location including locating the between the point and the triangle by calculating the acreage of triangle and combining the center of gravity of triangle with the relationship between the directed line segment and the point. Consequently, the search numbers of the triangles are decreased and the location of triangles is low time-consuming. While optimizing the local Delaunay triangulation network, it stores all the adjusted edges and vertices by Delaunay Quadtree, which can search effectively. Experimental result proves that applying the algorithm into the 3D terrain simulation makes good natural looking, and the time complexity of whole algorithm is low.
-
Computer Engineering. 2012, 38(22): 291-292. https://doi.org/10.3969/j.issn.1000-3428.2012.22.073The traditional modeling technology of dental restoration is often difficult to meet the requirements of precision. Therefore, this paper proposes a 3D coordinate measurement method based on line structured light. It calibrates the camera, caculates the light plane equation by least squares method, and uses a motorized translation stage and rotary stage scan method to get 3D surface data, which avoids the intrinsic and extrinsic parameters calculation of the camera. Experimental result indicates that the 3D model after reconstruction can meet the precision requirement.
2024
Volume 50 No 3
Volume 50 No 3
(Started in 1975, Monthly)
Sponsored by：
East-China Institute of Computer Technology Shanghai Computer Society
Edited & Published by：
Editorial Board of Computer Engineering
Chief Editor：Jiang Bo
Executive Chief Editor:
Hu Bing
Tel：021-67092217
E-mail: ecice06@ecict.com.cn
CN: 31-1289/TP
ISSN: 1000-3428
Sponsored by：
East-China Institute of Computer Technology Shanghai Computer Society
Edited & Published by：
Editorial Board of Computer Engineering
Chief Editor：Jiang Bo
Executive Chief Editor:
Hu Bing
Tel：021-67092217
E-mail: ecice06@ecict.com.cn
CN: 31-1289/TP
ISSN: 1000-3428