Applied Mechanics and Materials Vols. 543-547

Abstract: In face image with complex background, the CbCr skin color region will have offset when considering the illumination change. Therefore, the non-skin color pixels which luminance is less than 80 will be mistaken as skin color pixels and the skin color pixels which luminance is greater than 230 will be mistaken as non-skin color pixels. In order to reduce the misjudgments, an improved skin color model of nonlinear piecewise is put forward in this paper. Firstly, the skin color model of non-piecewise is analyzed and the experimental results show that by this model there is an obvious misjudgment in face detection. Then the skin color model of nonlinear piecewise is mainly analyzed and is demonstrated by mathematics method. A large number of training results show that the skin color model of nonlinear piecewise has better clustering distribution than the skin color model of non-piecewise. At lastly, the face detection algorithm adopting skin color model of nonlinear piecewise is analyzed. The results show that this algorithm is better than the algorithm adopting skin color model of non-piecewise and it makes a good foundation for the application of face image.

Abstract: The application scheme of Campus Smart Card System is analyzed, and its design aim and development principle of the functional modules are introduced. campus smart card system on 3G technology which is one of the important parts in campus digitalization. It offers comprehensive data collection and excellent information sharing environment. Its realization will promote the information management of the school and increase construction process of digital campus.

Abstract: Edge detection plays an important role in computer vision and image processing. Fractal and Fuzzy theory show significant effect in the edge detection and have attracted much attention. Compared with traditional edge detection methods, this paper proposes a Fuzzy Box-counting Dimension Method (FBDM). This algorithm introduces the pre-judging mechanism to improve the speed of image segmentation, and the self-adaptive dimension threshold and the voting mechanism under multi-windows to improve the accuracy of the determination of edge points. Finally, closest principle is used to clear edge and reduce noise. Experimental results show FBDM can improve the precision of image edge detection effectively without pretreatment, and it has a very superior de-noising performance.

Abstract: This paper proposes a real-time and accurate human detection method base on a new Gradient CENTRIST feature descriptor. Firstly, the feature can characterizes not only local human appearance and shape but also implicitly represent the global contour. Secondly, it does not involve image pre-processing or feature vector normalization, and it only requires steps to test an image patch. Our main contribution is that a more reliable feature descriptor is found, which can get a better human detection. The experiments on the INRIA pedestrian dataset demonstrate that the detection performance is significantly improved.

Abstract: The focus of this paper is on Metric Learning, with particular interest in incorporating side information to make it semi-supervised. This study is primarily motivated by an application: face-image clustering. In the paper introduces metric learning and semi-supervised clustering, Boost the similarity metric learning method that adapt the underlying similarity metric used by the clustering algorithm. we propose a novel idea of learning with historical relevance feedback log data, and adopt a new paradigm called Boost the Similarity Metric Method for Face Retrieval, Experimental results demonstrate that the unified approach produces better clusters than both individual approaches as well as previously proposed semi-supervised clustering algorithms. This paper followed by the discussion of experiments on face-image clustering.

Abstract: We present a novel moving shadow detection method using logistic regression in this paper. First, several types of features are extracted from pixels in foreground images. Second, the logistic regression model is constructed by random pixels selected from video frames. Finally, for a new frame in one video, we take advantage of the constructed regression model to implement the classification of moving shadows and objects. To verify the performance of the proposed method, we test it on several different surveillance scenes and compare it with some well-known methods. Extensive experimental results indicate that the proposed method not only can separate moving shadows from moving objects accurately, but also is superior to several existing methods.

Abstract: The paper studies the problem of communicating message secretly in the network performing random linear network coding, where the network internal nodes are allowed to randomly mix the incoming packets and then forward. The paper proposes HoNet, an end-to-end homomorphic encryption that is theoretically proved secure against network adversaries who can fully access the information of network coding schemes and eavesdrop every network transmission. A direct payoff of homomorphic encryption is that network nodes could on-line decrypt (or re-encrypt) the mixed ciphertexts, which significantly increases network throughput in scenarios such as peer-to-peer networks and satellite systems. In particular, HoNet addresses the two main challenges faced by the traditional homomorphic encryption schemes for point-to-point transmissions:high computational overhead and throughput loss-rate. To be precise, HoNet possesses linear time nodes complexity and asymptotically zero loss-rate.

Abstract: Propose a preamble detecting Algorithm in the processing of MOD-5 signal, after researching in the feature of MOD-5 interrogating preamble pulse. This paper resume the MOD-5 working principle, demonstrate the algorithm flow and simulation parameters. An analysis of the simulation result had been done at last. This algorithm can provide the ability of noise suppression and precise timing. On that basis, it can be the foundation in IFF (Identification Friend of Foe) signal processing with the purpose of increasing successful identification probability.

Abstract: In this article, aiming at the flaws in traditional Mean-shift algorithm for object tracking, we improve the traditional algorithm based on the fusion of Grayscale and Gradient using a layered approach. This algorithm can describe the object more exactly by using Grayscale and Gradient dual tracking factors in object tracking; And it makes up the flaws in traditional Mean-shift algorithm for information expression of spatial location by using layered approach and spatial histogram in describing object. Experimental results show that modified Mean-shift algorithm is superior to traditional Mean-shift algorithm in object tracking with better Robustness and precision, and it can suit deformation, spin, interference, shield and more complex tracking problems.

Abstract: Currently, researches on the preparation and measurement of quantum state between two-dimensional dots in quantum network are mainly dot-to-dot or dot-to-multi-dot transmission schemes. The existing experimental and theoretical work is relatively scattered, and the overall knowledge is incomplete. The article proposes the preparation of multiple quantum entanglement state, the measurement and classification of multi-state and multi-body quantum entanglement, the design of the preparation and measurement scheme of quantum state between any two one-dimensional dots and proposes the transmission scheme of quantum state between two-dimensional dots in quantum network. Theoretical guidance is provided for relevant experiments through deep study on the problem.