Recent Patents on Computer Science - Volume 10, Issue 4, 2017

Background: In civil aviation information monitoring system, optimization of image recognition is applied to promote monitoring of and control over passenger mobility. The traditional image recognition by video surveillance cannot effectively detect abnormal behaviors or explosives, as described in various patents. Method: In this paper, the author proposes a method for the optimization of surveillance image recognition in civil aviation airport based on contourlet domain edge detection. Firstly, an overall model of surveillance image recognition is established and statistically significant probability analysis and other data integration methods are employed to realize comprehensive treatment of visual images. In order to enhance the light-and-shade contrast of moving regions in the images and make images smoother, we must evaluate edge position information of surveillance images, extract the lowfrequency parts and signals to enhance contrast and promote image recognition capability. Results: Simulation experiment proved that this method produced better image recognition results and could effectively detect abnormal behaviors and violent terrorists. Conclusion: It is a superior algorithm, which is of great importance to ensure the safety of airports.

Background: The rough set theory is a powerful tool to deal with imprecise and incomplete information in the field of data mining. As a core content for rough set theory, attribute reduction aims at removing redundant data and drawing the minimum attributes while maintaining indiscernibility relation. However, traditional rough set theory is available for classical example which has disadvantages of time-consumption, large storage and low recognition accuracy. In this paper, we focus on an attribute reduction based on particle swarm optimization (PSO) to overcome the drawbacks of traditional rough set theory. Firstly, this paper reviews some important concepts of rough set and particle swarm optimization. Then, we establish the model of attribute reduction based on particle swarm optimization. Finally, the proposed method is applied to actual oil logging data, and the reduction results are recognized by Relevance Vector Machine (RVM) and Second Order Cone Programming-Relevance Vector Machine(SOCP-RVM). The experimental results show that the proposed method is efficient and has high recognition accuracy. Methods: Recent publications and patent databases are reviewed to find extraordinary and innovative attribute reduction algorithms for reducing time consumption and accuracy. Results: Two methods which are RVM and SOCP-RVM are applied to recognize the attribute reduction results. The results show that well in 993-997m, 1045-1152.5m and 1236-1255m depth are main oil-layers, the rest are dry-layers (Fig. 2 shows that oil-layers are 995-997m, 1045-1152.5m and 1241.5-1255m; Fig. (3) shows that oil-layers are 993-996m, 1055.5-1143m and 1236-1251m). The recognition results are consistent with the actual oil test results. Conclusion: A novel PSO-Based Attribute Reduction of Rough Set is proposed, and apply it to oil well to deal with actual and complex data. Experimental results show that the proposed algorithm can get effective reduction sets, and the recognition results with high accuracy can be obtained in actual well by using RVM and SOCP-RVM algorithms, which are consistent with the actual oil conclusion. It indicates that the proposed attribute reduction based on PSO is practical and viable, and the reduction results are efficient.

Background: Traditional network management solutions suffer from scalability issues and delayed fault reporting when monitoring large network topologies, as described in various patents. IntelliMon is an intelligent polling engine which takes into account network baselining statistics, indicative of the past behavior of the network and utilizes this information to dynamically formulate its polling strategies to achieve faster detection of network faults, while regulating the amount of network management traffic that it generates. This leads to lower Mean-Time-to-Detect (MTTD) for network faults compared to traditional network management solutions, which employ static pre-configured polling strategies. Method: Network baselining statistics are used to formulate dynamic polling strategies for nodes or areas of the network which are prone to faults. Results: Early simulation results of up to 5000 node network show that IntelliMon achieves significantly lower MTTD, specifically 14-32% lower than traditional polling engines under different scenarios. Conclusion: Heuristics combined with hard-statistics derived from network baselining are effective in formulating adaptive polling strategies that are able to detect network faults quicker especially for large topologies.

Background: Face recognition has a very important application value in the field of information security as an important method of bioinformatics identification. There are recent patents that discuss a human face similarity recognition method and system. It has also faced the problem of complex feature space and the very large amount of data, which make face recognition one of the most challenging and most academic research topics. Method: In order to solve the problem of the lack of prior knowledge in the face recognition algorithm based on the traditional convolution neural network, this paper improves the traditional convolution neural network from the two aspects of feature extraction and classification recognition, and proposes a new method-face recognition algorithm based on the sparse representation of denoising autoencoder convolution neural network, SRDAECNN. Results: Extensive experiments are performed on LFW, ORL, YALE and other face database. The experimental results show that our proposed face recognition algorithm has high accuracy. Conclusion: The model combines the advantages of the convolution neural network and sparse representation- based classifier, which can overcome the problem of incomplete feature extraction due to the random initialization of convolution kernel, and introduce sparse representation algorithm on classification recognition to enhance the recognition effect.

Background: Cloud computing is becoming prominent as it makes use of a model, in which consumer must pay according to its usage, as described in various patents. The user pays as per his demand and requirement. There are several issues faced by Datacenters for efficient scheduling of the workload. Task implementation failure is a very common property of cloud computing environment and is not given much attention in different scheduling techniques. In this research article, we propose a technique which takes the heed of defects using the concepts of autonomous computing. Methods: To assure the services related to quality to the consumers, an important task is to plot the available resources according to the jobs. Clustering is used where the events are logged and the ranges for CPU, Memory and Bandwidth are set as well. Fault tracing methodology is used with the help of which, the violations are checked, and requests are scheduled according to the results obtained by comparing the request with the cluster data. Result: In our proposed model (FSBD), we have tried to overcome the shortcomings of the existing techniques. The damage caused to the Service level agreement (SLA's) is less and at the same time, execution time is reduced and performance is enhanced. The experimental result shows that the computing which is sensitive towards the faults supports flexible contingency which is favourable in terms of lesser SLA violation, better time to respond (up to 4.46 ms) and shorter execution time. The proposed approach, when compared to the traditional approach of fault aware pattern recognition, showed better results in terms of forbearance of faults. Also, the number of failed cloudlets is significantly lesser in FSBD (4.9%) as compared to the traditional Round Robin (40%) approach. Conclusion: As is evident from the results shown, we can conclude that faults are able to cause huge damage to SLA's and lead to a lower performance in cloud computing. Further, we compared a system having no fault with the system having faulty behaviour to quantify the damage. After justifying the seriousness of the damage caused due to the fault, the proposed model recognizes the pattern of the behaviours of each component of the virtual machine, thereby identifying the problematic Virtual machine (VM) in the system. Post identification, very little number of requests is being allocated to the faulty VMs to keep SLA intact. Experiments conducted for validating the architecture clearly showed the effectiveness of the scheme.

Background: Due to the large network scale, nowadays, it is hard to get extensive data from online social networks (OSN). Moreover, a large number of social nodes and links have made network data analysis a time-consuming task. Therefore, to sample the large-scale online social networks and restore the topological properties of original network become a problem. The purpose of this paper is to study an unbiased sampling method that can extract a representative sample from the social graph. Methods: We propose an improved algorithm based on MHRW, called Unbiased Delay sampling (UD algorithm). Then we compare it with some recent patents on sampling method to evaluate our method. Results: Different sample methods extract subnet with different topological properties. We find that UD can adapt to all kinds of different network connectivity. On the one hand, UD has a better degree distribution when the sample does not consider repeated nodes; on the other hand, UD algorithm can reduce the probability of reiterated nodes selected to sample and improve the ability of network discovery. Conclusion: We get the first, to the best of our knowledge, unbiased sampling method which has a good degree of distribution when the sample set does not have duplicate nodes. More specifically, we add parameter α to sampling process, and the value of α can control the repetition rate of the sample set.

Background: Problems of cloud user privacy leaks have already hindered the further development of cloud computing, as described in various patents. Therefore, the comprehensive analysis and measurement of cloud computing privacy risks factors is an effective way to control and identify risks. However, the researches about this issue have not yet been discovered, so the relevant issues have been focused in this paper. Methods: This paper combines Analytic Hierarchy Process(AHP) with Information Entropy theory to identify and measure the privacy security risks of cloud computing. Results: This paper constructs a privacy risk attribute model of cloud computing with 20 privacy risk factors, based on the model, a cloud computing privacy risk measurement model has been designed. Conclusion: In order to effectively identify the defects and weakness on privacy protection, this paper can calculate the value and level of privacy risk factors of the attribute model by measuring model. The measurement results can provide a reference for cloud service providers to protect the privacy of users. Then the correctness of the model is verified by relevant simulation experiments.

Background: The supply chain is an integrated network, which is generally composed of suppliers, manufacturers, distributors, retailers and ultimate users. In the supply chain system, each member has different effect on the system operation. The importance measure can be used to help the managers to recognize the weak components of the system. In the related patents of supply chain networks, they seldom consider the importance analysis. Methods: Based on the reliability analysis of the supply chain, this paper introduces the importance measure to analyze the influences of supply chain members on the entire system. Then the importance values of different supply chain members are compared, and weaknesses recognition of the supply chain is analyzed. Results: According to the importance values of different members in supply chain networks, when the members have the same shortage rates, the importance in the series position is larger, and the manufacturer has the most effect on the supply chain systems. Conclusion: During the whole period, the ranking of Birnbaum importance is the manufacturer, supplier and distributor, and retailer. These are helpful to increase the operational efficiency of the supply chain and provide effective methods for improving the supply chain management.

Background: Biometric testing concerning face recognition makes it hard to solve due to the inaccuracy problem. Alongside the present progress in many technological fields, there are still different critical issues that affect the performance of real-time face recognition systems. Methods: Recent publications and patent databases related to face recognition are reviewed to find the best classifier of face recognition. In addition, these publications and patents are concerned to improve the face recognition system especially its real-time performance. In this paper, we introduce a new multi-agent system that will improve the face recognition system especially its real-time performance. Results: Face recognition done using multi-classifier (K-NN, NN, and CART) and multi-agents incorporated agent with a multi-feature approach. Five types of agents are used in our experiments namely; information agent, preprocessing agent, classifier agent, headquarters agent, and communication agent. The experimental results showed that the recognition rate improved. Face recognition accuracy up to 99.5% interpreted as 1.5 seconds in threading mode, and 1 second in distributed mode. Conclusion: By using multiple agents, the recognition processing time was improved. The use of multi-feature extraction turned out to be more efficient in the recognition accuracy. The proposed model proved to be robust in time using distributed mode execution for the classifier agents group. In addition, tapping the issue of distributed vs. threading mode distribution of agents makes a great link to the upcoming challenges of nowadays-modern sciences.

Background: Recently, with the increasing shortage of energy, researchers paid more attention to wind power, photovoltaic, hydroelectric and other renewable energy technologies. As one of the most important renewable energy sources, hydro energy is widely used, as described in the various patents. Meanwhile, the scholars put forward the concept of micro-grid which is a controllable unit consisting the generator, load, energy storage device and control device, micro-hydro generator as one of the most important micro sources has been getting more and more attention. So, in this paper, a grid-connected topology circuit of micro hydropower system based on self-excited induction generator is proposed. Methods: Micro hydropower system based on self-excited induction generator drives the turbine by the fixed amount of hydraulic potential difference, and then drives the induction generator to generate stabilization power. This controller includes the AC/DC module and DC/AC module. The AC/DC module uses hysterics PWM current control, while DC/AC module uses voltage outer loop, and current inner loop of double closed-loop control. Results: In this paper, a resistive load which can be varied from 0 to 4.5kW in steps is used as consumer load and the variety of grid voltage and grid current when the load is changed. When the system starts running, the grid voltage can be tracked by grid current with the same frequency and phase, the power factor is 1. And the current has a small impact on voltage. the waveforms of grid voltage and grid current when the load is added from 0W to 300W. The power of generator is used by load, so gird current is reduced. Conclusion: A topology structure of micro hydropower based on self-excited induction generator has been proposed. It has been presented in an economical and energy-efficient way. A comparative study of both the rectifier is uncontrolled and controlled in the system has been demonstrated by simulation. The system of excited capacitance has been proposed which gives a steady voltage. The experimental investigation shows that the grid voltage can be tracked by grid current with the same frequency and phase, and the result of the grid is an excellent option for micro-hydro applications.