Applied Mechanics and Materials Vols. 602-605

Abstract: It is very important to protect the safety of the human head with helmet. Traditional detection for helmet wearing mainly relies on manual approach, which was more subjective that a missing condition may happen caused by fatigue and other factors. Owing to this situation, this paper proposed a method for automatic detection of operator without helmet in real-time. Firstly, Gaussian model for background subtraction is used to detect moving target. Secondly, HOG feature extraction can be used to classify the human target from vehicle. Then, a color feature extraction algorithm is proposed for helmet recognition. The algorithm has been applied into the real time monitoring system and verified with higher accuracy.

Abstract: By monitoring and controlling the temperature and stress of concrete in the process of mass concrete casting, the temperature stress can be reduced effectively. In response to problems of traditional temperature and strain monitoring method, a new kind of distributed temperature and strain monitoring system based on mobile-agent and GPRS had been proposed. And also this system design a data integration middleware based on mobile-agent which verify the validity of the middleware through the experiment on Genomote nodes of network. Through GPRS network, remote data monitoring center receives data packets of temperature and strain of the concrete. The real-time display function of temperature and strain parameters was achieved by PC software programmed by Labview. The experiment show that the temperature acquisition value and the standard temperature error is within the scope of the permit (<1°C) and the system is stable, reliable for the needs of the remote monitoring.

Abstract: Shadows may cause many problems in computer vision, such as object recognition, image segmentation and video surveillance. In this paper, we present a new method to detect cast shadow in a single outdoor image. We build up an illumination model to explain the process of shadow formed, and through this model we introduce some useful features. The regions for extract features are acquired through canny edge detector, after a series of morphological operations. Then we use SVM classifier with a multi-kernel model to train these features for shadow region classification. Our results show that edges of shadow images can be detected effectively with our methods.

Abstract: To solve the problems caused by the complex failure mechanisms in hydraulic brake failure systems and the uncertain linkages between fault type and fault symptoms, a method of Bayesian network (BN) is offered for failure diagnosis. A statistical strategy is adopted there in this algorithm on the rule base provided by many experts to discard the weak causal relationship rules while keep the relatively strong ones during the BN structure learning, and a BN-based model of layered architecture is established for fault diagnosis of hydraulic braking failure systems. Experimental data analysis shows that the Bayesian network fault diagnosis model has higher accuracy than fuzzy logic diagnosis method, effectively solving the uncertainties in fault diagnosis.

Abstract: A novel visual object tracking method for color video stream based on traditional particle filter is proposed in this paper. Feature vectors are extracted from coefficient matrices of fast three-dimensional Discrete Cosine Transform (fast 3-D DCT). The feature, as experiment showed, is very robust to occlusion and rotation and it is not sensitive to scale changes. The proposed method is efficient enough to be used in a real-time application. The experiment was carried out on some common used datasets in literature. The results are satisfied and showed the estimated trace follows the target object very closely.

Abstract: The parallelism between an industrial camera and a servo motion direction is corrected with the help of image measurement to shadowed geometric contours. Then a nearly orthogonal angle between XY servo motion directions is obtained according to an inherent geometry relationship in contours. The installation error of a PCB in platform is compensated based on automatic multi-spot imaging finally. An experimental prototype was built while the PCB alignment was implemented on a lot of samples according to the method introduced above. It proves that the developed immediate alignment method as well as its specific embodiment fulfills the requirement of positioning accuracy in the initial design.

Abstract: For the non-stationary characteristics of rotating machinery fault vibration signal, proposed a fault diagnosis method that based on ensemble local mean decomposition (ELMD) to extract fault feature, and fuzzy C-means clustering (FCM) to perform the fault identification. ELMD method can effectively solve the problem of aliasing modes in LMD. Firstly, decomposing the fault vibration signal by ELMD, PF components were obtained in which the initial feature vector matrix, The PF components compose a initial feature vector matrix, and do singular value decomposition, using the singular value decomposition feature vector as the fault characteristic vectors. Finally, using FCM clustering as a fault classifier. Achieved the identification of different fault types. Experimental results show that this method can effectively achieve the bearing fault diagnosis.

Abstract: For the real-time measurement of transmission performance in the real-time optical OFDM transmission system, realization of BER test with an optimum method by utilizing FPGA, which means to complete the calculation of BER inside transceiver, become essential and efficient. An experiment setup demonstrate the dramatic real-time property and effectiveness in testing the optical OFDM transmission system.

Abstract: ShellCode injections with malicious JavaScript code in documents are becoming more prevalent and dangerous. However, the existing methods have some limitations in detecting this kind of attacks. In this article, we explore the detections of malicious documents and propose an approach of detecting malicious documents that contains JavaScript ShellCode. In our approach, we provide an impact factor which represents the reliability of the document being malicious. We use both static detections and dynamic detections and then combine the results of the two different methods. Therefore, we can get an acceptable overhead and make the detection immune to obfuscation. We have implemented a proof-of-concept prototype of the detection system on a Linux platform. We also have evaluated the accuracy and the performance overhead on the test platform. The results show that the system reports very few faults with an acceptable overhead.

Abstract: Object tracking across non-overlapping views is a hot and important research topic in compute vision. In this paper, a novel method to track the interested object continuously across non-overlapping cameras is presented. This challenging task is taken as two sub-problems: single camera object tracking and object matching across disjoint cameras. An object tracking algorithm which improves Tracking-Learning-Detection (TLD) algorithm by adding background extraction and Kalman filter is presented to deal with the first problem. A new object matching algorithm based on the fusion of global features and local features at the assistance of 3D GIS is also introduced for object matching across disjoint cameras. The proposed approach does not need a training phase and inter-camera calibration. Experiments are carried out on real world videos to validate the proposed approach.