Search

In this paper, a technique is presented for the fusion of Panchromatic (PAN) and low spatial resolution multispectral (MS) images to get high spatial resolution of the latter. In this technique, we apply PCA transformation to the MS image to obtain the principal component (PC) images. A NSCT transformation to PAN and each PC images for N level of decomposition. We use FOCC as criterion to select PC. And then, we use the relative entropy as criterion to reconstruct high-frequency detailed images. Finally, we apply inverse NSCT to selected PC's low-frequency approximate image and reconstructed high- frequency detailed images to obtain high spatial resolution MS image. The experimental results obtained by applying the proposed image fusion method indicate some improvements in the fusion performance.

In order to get the change detection image.An unsupervised change detection algorithm for multi-temporal satellite image based on NSCT (non-subsampling contourlet transform) and k-means clustering is proposed in this paper. For each pixel in the log-ratio image, multi-scale and multi-direction feature vector is extracted by NSCT and the reconstruction of the log-ratio image is obtained. The threshold is produced by using the k-means clustering algorithm and can distinguish between the unchanged and the change region. Finally, the change detection map is achieved. Some satellite images are used to verify the proposed method and the results shows that it has a higher stability and accuracy against Gaussian and speckle noise than traditional algorithms.

In cellular mobile communication network, for using the limited available spectrum to meet the increasing demand of customer, it is important to scheme frequency source by applying some optimized algorithm. To the question, an improved discrete harmony algorithm is proposed, particle location update strategies of the particle swarm optimization algorithm is introduced, the probability of harmony memory size and the probability of pitch adjusting rate are adjusted dynamically, which improve the global optimization stability of the search results, enhance the universality and robustness of the algorithm, and improve the convergence rating and the convergence speed. Simulation results show that the improved algorithm applied to solve the problem of frequency assignment has achieved good results.

In order to get better control of the chaotic randomness and improve the convergence rate of Transient Chaotic Neural Networks(TCNN),Effective improvement is proposed in this study.By modifying the Incentive function with composite sine sigmoid function and Time-varying Gain are adopted in the function, and two-stage annealing mechanism is applied in the annealing process of the dynamic equations.Empirical results show that combination of the two aspects has a better control of the two processes of the chaotic state,improved algorithm obtains a wealth of chaotic characteristics. Experiments on the channel allocation problem indicate that the modified algorithm can better solve this kind of Combinatorial optimization problems. Results are compared with other prior algorithms such as basic TCNN and only subsection of TCNN.The comparison shows that the performance of the algorithm has been greatly improved, local optima is avoided, and the convergence speed is improved more than 14%.

A new method for image denoising based on the free distributed hypothesis test threshold (FDR) and the non-sub-sampled contourlet transform(NSCT) is proposed in this paper. This method firstly acquires the free distributed false discovery rate hypotheses test in statistics to set the threshold in the NSCT domain, and then removes the noise through soft threshold function, which doesn’t depend on the length of signal. The experimental results show that the proposed method can more effectively reduce Gaussian noise and improve the peak value signal-to-noise ratio in the remote sensing image; Meanwhile, this method utilizes the shift invariant of NSCT transform to inhibit the pseudo Gibbs distortion effect, and integrally preserves the texture and edge etc.. details’ information of the image, thus obviously ameliorate the visual effect of the image.