Entity

Time filter

Source Type


Chen C.,Xidian University | Chen C.,Chinese Aeronautical Radio Electronics Research Institute | Liu G.,Xidian University
Chinese Optics Letters | Year: 2012

An effective image splicing algorithm based on phase correlation and speeded-UP robust features (SURF) operator is proposed which can sort the disordered sequence and stitch them into a super viewing field image without any human intervention. Phase correlation in frequency domain is used for images sorting and region of interest (ROI) estimation, and guiding features extracting and matching in spatial domain by SURF operator and bidirectional best bin first (BBF) strategy. The experimental results demonstrate that this algorithm not only can deal with the input images with translation, rotation and scale changes, but also outperforms the pre-existing methods on the aspect of repeatability, efficiency and accuracy. © 2012 Chinese Optics Letter. Source


Niu W.,Northwestern Polytechnical University | Wang G.,Chinese Aeronautical Radio Electronics Research Institute | Zhai Z.,Northwestern Polytechnical University | Cheng J.,Xian Institute of Applied Optics
Applied Mechanics and Materials | Year: 2012

Recently, the dominating difficulty that fault intelligent diagnosis system faces is terrible lack of typical fault samples, which badly prohibits the development of machinery fault intelligent diagnosis. Mainly according to the key problems of support vector machine need to resolve in fault intelligent diagnosis system, this paper makes more systemic and thorough researches in building fault classifiers, parameters optimization of kernel function. A decision directed acyclic graph fault diagnosis classification model based on parameters selected by genetic algorithm is proposed, abbreviated as GDDAG. Finally, GDDAG model is applied to rotor fault system, the testing results demonstrate that this model has very good classification precision and realizes the multi-faults diagnosis. © (2012) Trans Tech Publications, Switzerland. Source


Niu W.,Northwestern Polytechnical University | Wang G.,Chinese Aeronautical Radio Electronics Research Institute | Zhai Z.,Northwestern Polytechnical University | Cheng J.,CNGC the 205 Institute
International Journal of Digital Content Technology and its Applications | Year: 2011

Fault prediction is of great importance to ensuring weapon equipment safety and reliability. Usually the data for fault detection and prediction of weapon equipment have feature like small samples, although the current main fault prediction methods have achieved certain success in practical application, they all fall short in some aspects. For chaos of weapon equipment fault data, based on rough sets and support vector machine modeling theory, an optimal least square support vector machine prediction method is proposed. Firstly, redundant information in time series is removed by rough sets. secondly, time series after reduction is prediction by support vector machine. The data of a certain aeroengine are taken as an example for prediction and analysis, and the results show that the model simplifies complexity of modeling and has high prediction precision, which in turn validates its availability. Source


Lv G.,Northwestern Polytechnical University | Hu S.,Chinese Aeronautical Radio Electronics Research Institute | Lu X.,Northwestern Polytechnical University
ICALIP 2014 - 2014 International Conference on Audio, Language and Image Processing, Proceedings | Year: 2015

This paper introduced the semi-continuous Hidden Markov Model (HMM) and proposed a novel Dynamic Bayesian Network (DBN) model for dynamic speech emotion recognition. The former reduces the training complexity caused by mixture Gaussians by sharing the Condition Probability Densities (CPDs) of Gaussians among the states, and the latter adds a sub-state layer between state and observation layer based on traditional DBN framework and describes the dynamic process of speech emotion in detail. Experiments results show that average emotion recognition rate of semi-continuous HMM is 4% and 10% higher than those of classical HMM and Mixture Gaussian HMM respectively, and average emotion recognition rate of the three-layer DBN model is 11% and 8% higher than those of traditional DBN model and semi-continuous HMM. © 2014 IEEE. Source


Gao J.,Beihang University | Zhang Y.,Chinese Aeronautical Radio Electronics Research Institute
Proceedings - 2013 International Conference on Virtual Reality and Visualization, ICVRV 2013 | Year: 2013

The PnP (perspective-n-point) problem is very important in pose estimation technique based on computer vision. Aiming at this issue, an improved iterative solution is proposed. By the means of expressing the 3D point coordinates as a weighted sum of four control points, a system of homogeneous linear equations was established and then the optimized projections on the normalized image plane were obtained. The final estimation result was achieved by a relaxation-based iterative approach. Both simulations and experiments certify that the proposed algorithm can improve the computing accuracy and depress the image noise. Compared with other solutions to the PnP problem, the proposed algorithm can reduce the computational complexity while maintaining high precision with noise depression capability. © 2013 IEEE. Source

Discover hidden collaborations