Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-08-13 Cooperative journals: 《计算机应用研究》
Abstract: The clustering algorithm of based on grid density is difficult to determine the grid width and density threshold. In addition, the accuracy of the results is dissatisfied. Considering the problem above, this paper proposed an improved clustering algorithm. The better segmentation width of each dimension is calculated by the natural distribution information of the data set. According to the different density thresholds, the number of the noise is calculated. The noise curve is drawn. The best density threshold is obtained from the noise curve. Simulation results show that the improved algorithm can get better clustering results.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》
Abstract: To solve the problems of network congestion and load imbalance caused by elephant flow that carries large amounts of data in data center networks. This paper proposed an SDN based load balancing mechanism of elephant flow(EFLB) . When network load exceeded the threshold, the controller split the detected elephant flows down into multiple mice flows by Openflow feature and calculated dynamically the minimum load switch according to the collected network topology and link states to ensure load balancing. Experimental results show that the EFLB achieves network load balancing by improving the network throughput and link utilization compared with ECMP((equal-cost mulit-path routing) .
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-24 Cooperative journals: 《计算机应用研究》
Abstract: Electro encephalog ram(EEG) can reflect the thinking activity of the brain under different conditions, therefore, motor imagery recognition based on EEG has become a new research hot spot. To reduce the influence of low quality samples on the session-to-session transfer performance of CSP filter models and improve the recognition accuracy ratio, this paper proposed an incremental updating algorithm for CSP filter based on training sample evaluation. To start with, it used sample selection method to evaluate the quality of EEG data. Then it removed a set of training data corresponding to low recognition rate. Finally, it updated the CSP filter incrementally which designed by the optimized sample. In label environment, the motor imagery recognition of EEG signals reaches average accuracy of 80.92%. Compared with the traditional CSP method, the average recognition rate of the five subjects' testing sets increases by 5.4%, 5.6%, 1.5%, 8.6%, and 7.7%, respectively. The experimental results verify the effectiveness of the proposed algorithm.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-17 Cooperative journals: 《计算机应用研究》
Abstract: The palmprint enhancement algorithm and the palm vein enhancement algorithm in single near infrared palm vein palmprint fusion recognition algorithm can not highlight the palmprint structure and palm vein structure well. To solve this problem, this paper proposed an improved palmprint and palm vein fusion algorithm. First, it used the block model to remove the palm to get the structure of the pure palmprint, and defined a new membership function to make palmprint structure fuzzified, then it combined the fuzzy palmprint structure with the anti sharpening mask method to highlight the palmprint structure information. Secondly, it used edge detection weighted guided filtering to enhance the structure of the palm vein so that the palm vein structure was highlighted. Finally, it acquired the palmprint and palm vein images for adaptive weight fusion. The experimental results show that the recognition rate of the improved fusion recognition algorithm reaches 99.81%.