This paper proposed a theoretically efficient approach for face recognition based on principal component analysis (PCA) and rotation invariant uniform local binary pattern texture features in order to weaken the effects of varying illumination conditions and facial expressions. Firstly, the rotation invariant uniform LBP operator was adopted to extract the local texture feature of the face images. Then PCA method was used to reduce the dimensionality of the extracted feature and get the eigenfaces. Finally, the nearest distance classification was used to distinguish each face. The method has been accessed on Yale and ATR-Jaffe face databases. Results demonstrate that the proposed method is superior to standard PCA and its recognition rate is higher than the traditional PCA. And the proposed algorithm has strong robustness against the illumination changes, pose, rotation and expressions.

Proposed a chaotic cellular neural network system, in-depth study of the system dynamic characteristics through theoretical analysis and simulation, to obtain the system Lyapunov exponent spectrum, also designed the hardware circuit, and get the physical implementation of chaos system through the FPGA, observing all kinds of chaotic attractor. The experimental results are fully consistent with the Matlab simulation results, thus this physical implementation is correct and feasible.

Based on two hyper-chaotic recurrent neural networks, a new image encryption scheme is presented in this paper. In the encryption scheme, the shuffling matrix is generated by using a Hopfield neural network, which is used to shuffle the pixels location; the diffusing matrix is generated by using a cellular neural network, which is used to diffuse the pixels grey value by OXRoperation. Finally, through numerical simulation and security analysis, the effectiveness of the encryption scheme is verified. Duo to the complex dynamical behavior of the hyper-chaotic systems, the encryption scheme has the advantage of large secret key space and high security, and can resist brute-force attacks and statistical attacks effectively.

In this paper, an tracking algorithm combing color and LBP texture features based on particle filter is proposed to overcome the disadvantages of existing particle filter object tracking methods. A color histogram and a texture histogram were combined to build the objects reference model, effectively improving the accuracy of object tracking. Experimental results demonstrate that, compared with the method based on single feature, the proposed method is highly effective, valid and is practicable.

It is important for identifying the correspondences of points in different images. A double epipolar lines approach based on linear searching was proposed to identify the corresponding points in three images. First the component expressions of a given point epipolar equation were derived from the known epipolar model. Then the linear searching method was utilized to solve the problem of no analysis solutions since the numbers of state variables in these expressions were more than those of independent equations. Finally the target was found in looking for the corresponding points. The simulation shows it is effective.