Face recognition has become a research hotspot in the field of pattern recognition and artificial intelligence. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are two traditional methods in ...
详细信息
Face recognition has become a research hotspot in the field of pattern recognition and artificial intelligence. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are two traditional methods in pattern recognition. In this paper, we propose a novel method based on PCA image reconstruction and LDA for face recognition. First, the inner-classes covariance matrix for feature extraction is used as generating matrix and then eigenvectors from each person is obtained, then we obtain the reconstructed images. Moreover, the residual images are computed by subtracting reconstructed images from original face images. Furthermore, the residual images are applied by LDA to obtain the coefficient matrices. Finally, the features are utilized to train and test SVMs for face recognition. The simulation experiments illustrate the effectivity of this method on the ORL face database.
Nowadays, gene chip technology has rapidly produced a wealth of information about gene expression activities. But the time-series expression data present a phenomenon that the number of genes is in thousands and the n...
详细信息
Nowadays, gene chip technology has rapidly produced a wealth of information about gene expression activities. But the time-series expression data present a phenomenon that the number of genes is in thousands and the number of experimental data is only a few dozen. For such cases, it is difficult to learn network structure from such data. And the result is not ideal. So it needs to take measures to expand the capacity of the sample. In this paper, the Block bootstrap re-sampling method is utilized to enlarge the small expression data. At the same time, we apply "K2+T" algorithm to Yeast cell cycle gene expression data. Seeing from the experimental results and comparing with the semi-fixed structure EM learning algorithm, our proposed method is successful in constructing gene networks that capture much more known relationships as well as several unknown relationships which are likely to be novel.
The prediction of protein structure that attracts many researchers to study is an NP-hard problem. In recent year, numerous algorithms are proposed. Although particle swarm optimizer has many advantages, the disadvant...
详细信息
This paper presents an analysis method to the global exponential stability for the delayed genetic regulatory networks with SUM regulatory functions. By transforming the kinetics of networks into a single delay differ...
详细信息
A kind of scale invariant features transformation (SIFT for short) operators on DWT domain are proposed for watermarking algorithm. Firstly, the low frequency of the image is obtained by DWT. And then the SIFT transfo...
详细信息
A kind of scale invariant features transformation (SIFT for short) operators on DWT domain are proposed for watermarking algorithm. Firstly, the low frequency of the image is obtained by DWT. And then the SIFT transformation is used to calculate the key feature points for the low frequency sub-image. Based on the chosen space's key points with moderate scale, a circular area as watermark embedding area is constructed. According to the research and final results, the novel digital watermark algorithm is proposed benefiting from the characteristics of SIFT's key points and local time-frequency of DWT. The algorithm not only has good robustness to resist on such operations as compression, shearing, noise addition, median filtering and scaling, but also has good inhibition to possible watermark fake verification.
A new image fusion encryption algorithm based on image fusion and DNA sequence operation and hyper-chaotic system is presented. Firstly, two DNA sequences matrices are obtained by encoding the original image and the k...
详细信息
A new image fusion encryption algorithm based on image fusion and DNA sequence operation and hyper-chaotic system is presented. Firstly, two DNA sequences matrices are obtained by encoding the original image and the key image. Secondly, using the chaotic sequences generated by Chen's hyper-chaotic maps to scramble the locations of elements from the DNA sequence matrix which generated form original image. Thirdly, XOR the scrambled DNA matrix and the random DNA matrix by using DNA sequence addition operation. At last, decoding the DNA sequence matrix, we will get the encrypted image. The simulation experimental results and security analysis show that our algorithm not only has good encryption effect, but also has the ability of resisting exhaustive attack and statistical attack.
In this paper, an image encryption algorithm based on combination of DNA coding with chaotic map is proposed. The proposed algorithm uses chaotic sequence generated by Logistic system to scramble the image pixels whic...
详细信息
In recent years many researchers are trying to combine wheeled movement with legged movement or hopping movement to strengthen the robots adaptability to environment. Hybrid robots are playing a more and more importan...
详细信息
In recent years many researchers are trying to combine wheeled movement with legged movement or hopping movement to strengthen the robots adaptability to environment. Hybrid robots are playing a more and more important role in our modern life. However, their controls are mostly complicated. So in this paper we present a new solution for the wheeled hopping robot. Based on the mechanical design, this robot has two movement modes: wheeled and hopping movement mode. While the center of gravity of the mechanism is under the wheels center, the wheeled hopping robot which is like a tumbler can keep its stability even when the wheels' velocity is zero. In this paper, we firstly introduce the description and design of the wheeled hopping robot. Then we explain how to choose the robot parameters and lastly we describe, how the robot through hopping, successfully traverses obstacles and ditches.
A 3D reconstruction technique was applied to acquire pavement surface texture since the traditional method can’t achieve the balance between accuracy and *** this study,a simple pavement texture detection platform co...
A 3D reconstruction technique was applied to acquire pavement surface texture since the traditional method can’t achieve the balance between accuracy and *** this study,a simple pavement texture detection platform consisting of a camera,lighted sources and tripod was developed basing on four sources photometric stereo *** the pictures captured by detection platform were analyzed by least square method to extract the normals of points in the pavement *** these normals were calculated,they were used to recover surface heights of pavements by tangent plane *** experimental result shows that the 3D model of asphalt pavement recovered by photometric stereo technique is efficiency to reveal the pavement surface topography both in macro-scale and micro-scale.
In this work, we further extended the face-based smoothed finite element method （FS-FEM） for modal analysis of three-dimensional solids using four-node tetrahedron elements. The FS-FEM is formulated based on the smo...
详细信息
In this work, we further extended the face-based smoothed finite element method （FS-FEM） for modal analysis of three-dimensional solids using four-node tetrahedron elements. The FS-FEM is formulated based on the smoothed Calerkin weak form which employs smoothed strains obtained using the gradient smoothing operation on face-based smoothing domains. This strain smoothing operation can provide softening effect to the system stiffness and make the FSFEM provide more accurate eigenfrequency prediction than the FEM does. Numerical studies have verified this attractive property of FS-FEM as well as its ability and effectiveness on providing reliable eigenfrequency and eigenmode prediction in practical engineering application.
暂无评论