Applied Mechanics and Materials Vols. 241-244

Abstract: As an important step of printed formula recognition system, formula extraction locates the formula fields on the layout images of printed documents, which influences the performance of formula recognition to a great extent. However, the errors of automatic formula extraction occur inevitably because of the complexity of formulas themselves and the layouts which the formulas situated. To solve this problem, this paper designed a post-processing method to correct the errors existing in the results of formula extraction algorithm according to relative layout knowledge. First of all, the geometrical features of various layout fields were employed to correct the extraction errors. Then, the syntax rules were used to check the boundary components of different kinds of areas in layouts to identify which field it should belong to. Finally, the formula area was adjusted according to above mentioned information.

Abstract: This paper mathematically described the group selecting problem of intermediate ordnance equipment support agency. Through taking maximizing support ability of the selected group and minimizing support ability differences of each equipment type as the objective function, taking a minimum support ability of equipment, staffing level of the selected group and support agency as conditions, optimization model was put forward. And the effectiveness of the model is validated by the example computation.

Abstract: To support code-switching bilingual teaching in computer major courses, an online e-Leaning system based on model-view-controller (MVC for short) architecture is designed and implemented using object-oriented analysis and design method. Here, Struts, Spring and Hibernate are chosen as development frameworks to implement MVC model. Google Translate API as a tool that automatically translates text from one language to another language is adopted to support web pages with multiple languages. Based on various teaching materials such as video or audio or flash, it is easy for students to finish self-study and for teachers to finish supplementary teaching. With this e-Learning system, it is convenient to arouse students’ interests of computer major courses, improving students’ English expression skills of speaking and writing and enhancing students’ international employment competence.

Abstract: In order to improve the accuracy and real-time performance of webpage filtering, a sensitive webpage filter based on multiple filtering was designed. Firstly, the URL is gained from IE browser’s address bar by BHO technology; Secondly, match the webpage text with sensitive vocabulary database using SMA algorithm; Finally, use the sensitive image detecting algorithm combing face detection, skin detection, skin text detection and classification to filter sensitive images in the webpage. The Simulation experimental results showed that the sensitive webpage filter can effectively intercept and filter sensitive webpages, meeting the accuracy and the real-time requirement of webpage filtering.

Abstract: Remote sensing (RS) image can be applied in many domains. Most research work on RS image retrieval is to meet the demand of professional user. However, there are demands for RS image that comes from non-professional users who propose the requests in natural language (NL) not filling in professional request forms. Some problems are needed to be solved to implement RS image retrieval based on NL user demand. The objective of this research was to propose a user demand semantic model to solve the problem of translation from NL user demand to value requirements. Based on plenty of materials investigated in application domains, the syntax and semantics of NL user demand was analyzed. Semantic relationship is summarized in terms of the semantic analysis. After that, a user demand semantic model is proposed and built with ontology. It can be conclude that the proposed semantic model may help to RS image retrieval based on NL user demand.

Abstract: As a Basic Transforming Operation between Time Field and Frequency Field, FFT Has Been Widely Used in Detection, Telecommunication, Signal Processing, Multimedia Communication Etc. the Implementation of the FFT Algorithms on FPGA Is Always the Hot Research Spots. in Order to Overcome the Shortcomings on the FPGA Resource Reusability Used in FFT Algorithm, this Article Discusses a New Configurable and High Efficient FFT/IFFT Soft-core Solution. the FFT/IFFT Soft-core Adopts Radix-2² Algorithm and Single-Path Delay Feedback (SDF) Pipeline Structure. its Configurable Factors Include: FFT/IFFT, FFT Points (2ⁿ, [3,12] ), Fixed-point Bit Width, Clock Delay of Complex Multiplier. the Design Takes FPGA Chip Stratix II EP2S130F780C4 as Hardware Platform, and the Complete Simulation and Synthesis Is Taken. the Maximum Operating Frequency Is up to 306.30MHz. if 300MHz Clock Frequency Used, 4096-point FFT Could Be Realized in 26.73us, and the Consumption of Memory Resources Is only 148Kbit. Compared with Altera FFT IP-core, Our FFT/IFFT Soft-core Has a Little Bit Longer Computing Time (0.6%). however, the LE Resource Consumption Is only 79% of Altera FFT IP-core. Platform, and the Complete Simulation and Synthesis Is Taken. the Maximum Operating Frequency Is up to 306.30MHz. if 300MHz Clock Frequency Used, 4096-point FFT Could Be Realized in 26.73us, and the Consumption of Memory Resources Is only 148Kbit. Compared with Altera FFT IP-core, Our FFT/IFFT Soft-core Has a Little Bit Longer Computing Time (0.6%). however, the LE Resource Consumption Is only 79% of Altera FFT IP-core.

Abstract: Wireless Mesh Networks are new type of multi-hop wireless network with high rate and strong reliability. Multipath routing is employed for load balancing, fault tolerance purposes in wireless mesh networks. With the more and more application need, multipath routing has gradually become an important topic to study. An algorithm of multipath routing algorithm based on dynamic source routing protocol(DSR) is proposed in this paper. This algorithm provides multiple QoS guarantee, such as bandwidth, delay and cost, achieves multipath load balancing and improves networks’ performance. The experimental results show that our algorithm achieves better QoS performance and more advantages over DSR and split multipath routing protocol(SMR).

Abstract: Traditional photo viewer can not provide 3D relationships of geometry scene between two photos. This paper presents a novel photo viewer using geometry method, and the only preparation work of our method is computing photos’ geometry information in advance. Our method first computed morphing matrixes from geometry information. We then constructed a camera motion using cosine function. Next, the morphing matrixes and camera motion were combined to draw a double-layer transition between two photos. Lastly, we demonstrated our photo viewer on a series of photos in campus. The results of a user study show that the experience of photo browsing is greatly enhanced with the proposed photo viewer.

Abstract: As an important value-added function for GIS service, GIS service chain provides the application foundation for reusing and automation composition GIS services. An important issue for realizing the value-added service is ensuring the operation results of GIS service chain meet the user needs, so it needs to verify the correctness of GIS service chain. This paper introduces the equivalence theory of Pi-calculus. A practical GIS service chain formal model is established using Pi-calculus. The validity of GIS service chain model and whether it meets the need are verified using the formal tools.

Abstract: The scheduling mechanism for cloud computing platform is described at first in this paper. Secondly the analysis of the B2C auto-negotiation process and the characteristics of its computation resource scheduling mechanism are given. Finally some computation resource scheduling strategies for B2C auto-negotiation in cloud computing platform are discussed.