Advanced Materials Research Vols. 1049-1050

Abstract: In this paper, we firstly analyzed the features or patterns of recognition errors in the text recognized by OCR.According to the results of Chinese egmentation and combining the analysis of san string, we detected recognition errors that may appear in the text through the application of proofreading rules or calculating Bi-neighborship of the words.At last, the experiment system would be tested and analyzed,and then we get the advantages and disadvantages of the method.

Abstract: Machine aided human translation (MAHT) for the abstract of patent texts is an important step to the deep processing of the patent data, where the terms have significant application value. This paper investigates the automatic term recognition (ATR), and proposes a new hybrid method based on two-phase analysis and statistic to generate English candidate terms. The segments including stop words were not simply discarded; instead, a rewriting method using beginning patterns, ending patterns, and inner patterns on the phase two was employed for the processing of the segments. In the meantime, generalized statistical measures were used for the evaluation of the candidates such as the generalized mutual information (MI), Log-Likelihood Ratio (LLR), and C-value to filter the low score’s candidate terms and to attain the intersection set of them. The experiments on the patent abstract texts extracted randomly show the availability of the method.

Abstract: This paper introduces suffix array method, the basic concepts, and cooperate with the longest suffix group public prefix array constructor, suffix array is constructed using multiplication algorithm, based on suffix array technology, has realized the extraction of multiple essay in this string of LCS, and then identify network hot topic, and then through the suffix array to assist in monitoring the essay in this cluster after all kinds of hot topic.

Abstract: In order to help department to make a decision whether the equipment need maintenance, some people trained the sample of characteristic parameter for riveting structure, and set up the model to recognize target by computer vision. However, we are difficult to find the research result about the affiliation between the characteristic parameter of the riveting structure and the model. In this paper, we make the image processing first, and use SVM (Support Vector Machines) algorithm to train the sample of characteristic parameter for rivet head. Finally, we research the affiliation between the characteristic parameter for the rivet head and the mathematical model, and test the accuracy of the model.

Abstract: Large-scale image recognition refers to giving computer the human visiual intelligent, in the massive data mode using computer to rrecognize the input image rapidly and exactly. In the process of recognition, the light, rotation and other factors will be the effects, meanwhile these noises will increase the difficulty of visual object recognition. How to recognize the large-scale image in the real scene and complex environment becomes a research topic. In order to recognize the large-scale image in real and complex envvironment and get a better recognition effect, this paper presents large-scale image based recognition algorithms with fusion of SIFT features and BP neutral network.

Abstract: An effective improvement method was put forward caused by the traditional Gaussian mixture model has poor adaptability to illumination mutation. When illumination mutation is detected, improved Frame difference could detect the foreground region and background region, and then adopts a new replacing update methods to the Gaussian distribution with the least weights of Gaussian mixture background models in different regions. The experimental results show that improved method makes Gaussian mixture model can quickly adaptive to the light mutation, and exactly detect the moving object.

Abstract: Community detection is an important approach to analyze and understand the organization or unit structure of the complex networks. By comparing the existing community detection algorithms, the label propagation algorithm (LPA) shows prominent operation speed and qualifies near linear time complexity. However, original LPA algorithm only uses the topological structure to guide the community detection process, failing to improve the quality of community detection when extra information offered. In this paper, we combine the prior information with topological structure to guide the community detection process. During the label propagation process, we proposed a new label update principle, making a node absorb its neighbor label information depending on the label distribution. The experimental results both on real networks and artificial networks show that the improved algorithm not only inherits the characteristic of rapid speed, but also improves the quality of community detection. Moreover, the improved algorithm still has the feature of near linear time complexity.

Abstract: Tracking moving objects with high accuracy is an important problem in fields of robot vision, video signal processing and pattern recognition. One of the widely used object tracking methods is Kalman filter. Conventional Kalman filter has a basic assumption that the noise obeys Gaussian distribution. However, in practice, the observed data also suffers from non-stationary sparse noise, which may cause performance deterioration for conventional Kalman filter. In this paper, we propose a robust Kalman filter based on convex optimization to remove not only the Gaussian noise but also such kind of sparse noise. Formulated as a convex optimization problem, the robust Kalman filter can be solved efficiently by Interior Point Method. Numerical results show that the proposed method is robust against sparse noise and achieves better performance while tracking objects under sparse noisy condition.

Abstract: Periodic signal is abundant in ship radiated noise. Its detection is an important step towards active control of ship’s acoustic signature. Based on power spectrum analysis and DEMON analysis, an approach to detect line spectrum and modulation spectrum of ship radiated noise is proposed. Periodogram method is used for power spectrum estimation. The separation of line and continuous spectrum, false positive removal, and line spectrum merging are conducted to improve the quality of detected characteristic line spectrum. Via DEMON analysis, steady physical characteristics including propeller’s rotational speed and number of blades are obtained. Finally, the effectiveness of the approach is demonstrated by simulation results.

Abstract: Automatic Guided vehicle (AGV) can free drivers from boring work. Previously AGV guided by black conduction band, as the color is single, it can only used for a kind of vehicle navigation.When there are several kind of vehicles need to be guided, the black conduction band becomes powerless. Color conduction bands can overcome this problem. Usually RGB image should be converted to other image space, and then identifies the bands,but it costs lots of time. This paper proposes a quick recognition method which can be used in RGB color space. It effectively improved the recognition speed, and realized real-time identification.