Papers by Keyword: Threshold

Abstract: The information security of wireless sensor networks is one of hot issues on the current research; this paper combines threshold key scheme with the management of security key sharing and group agreement and proposes dynamic key management scheme through clustering level key matrix of authentication mechanism from the cluster to node. The network is divided into different clusters, and the cluster head conducts the key authentication exchange and updates the key management with threshold key schemes, with no need of third authentication center, reduces the computing and communication costs, and promotes communication.

Abstract: The image segmentation of a robot binocular stereo vision system is the key issue in imaging processing. In this paper, the method of 2-D maximum entropy threshold image segmentation with chaos-particle swarm optimization algorithm is used to segment the images information collected by a robot vision system, and the algorithm is checked by a real robot binocular stereo vision system. The results show that more space information of an image is sufficiently used by the above method, and the running time is shortened by the algorithm. In addition, the image segmentation effect is better than other conventional methods.

Abstract: Otsu segmentation algorithm is one of the most successful algorithms in the field of image processing,However, it has drawbacks such as it is poor in dividing small object in images. This paper propose a improved Otsu algorithm based on weighted smooth histogram which can ensure the location of threshold point is more near to the valley point of the image and at the same time ensure the ensure between-class variance maximum.The results show that the improved Otsu algorithm can effectively accomplish segmentation for multimodal image, and get better segmentation results for images added Gaussian noise.

Abstract: A modified spreading dynamic model with feedback-mechanism based on scale-free networks is presented in this study. Using the mean field theory, the spreading dynamics of the model is analyzed. The spreading threshold and equilibriums are derived. The relationship between the spreading threshold, the epidemic steady-state and the feedback-mechanism is analyzed in detail. Theoretical results indicate the feedback-mechanism can increase the spreading threshold, resulting in effectively controlling the epidemic spreading.

Abstract: The central step of de-noising in wavelet domain is the threshold selection. To get the best de-noising result, the threshold should be selected according to the noisy observation. This paper presents a genetic adaptive threshold method, which gets the optimum threshold in the sense of least MSE by using an estimation function of the signals’ MSE function. The simulation results with standard wavelet test signals shows that the operation speed of the proposed method four times the traditional continuous searching algorithm, moreover, the optimum threshold calculated is more accurate and reasonable.

Abstract: How to identify and remove the noise of the dam monitoring data is an important work of dam safety-monitoring. The article makes use of a new compromise thresholding method to denoise the the dam horizontal displacement. Based on the de-noising data, the article uses the PSO-SVM to overfit and predict the dam horizontal displacement and compare the results with multiple regression.The result shows that the combination of the new compromise thresholding method and PSO-SVM has a better prediction accuracy.

Abstract: With the development of computer technology, digital image technology is also to grow rapidly. At present, the image processing has become a popular research field. Segmentation is a very important critical step in image processing tasks, which directly affects the subsequent analysis results. Cluster analysis belongs to the category of unsupervised classification and it is not depended on any prior knowledge of the category, so it usually serves as a basic tool to be used for the initial stage of image segmentation. In this paper, we propose a weighted fuzzy threshold algorithm based on fuzzy relations. This algorithm determines the fuzzy membership function according to the similar relationship between pixels, and emphasizes influence of sharpening features of Gaussian function. The new method overcome the drawbacks of clustering algorithms such as, sensitive to the initial value, need to pre-set number of categories. The numerical results show the effectiveness of the new method.

Abstract: Since the standard mean filter has some drawbacks, an improved mean filter is proposed in this paper to overcome them. In this improved algorithm, noise detection is used to find the noise which is caused by impulse wave in the image. Based on the result of noise detection, window size would be selected in the improved mean filter to minimize the influence of the noise. According to the result of computer simulation, it shows that the improved mean filter is much more efficient than the standard mean filter in eliminating the noise as well as in protecting the details of the image. It also shows its advantage in filtering speed.

Abstract: Lithium battery coating thickness is measured by sensor using laser triangulation. The algorithm based on threshold judgment and multi-scale wavelet is realized for noise high-frequency reduction of lithium battery film thickness systems. The result shows that compared with the unique multi-scale wavelet de-noising, the method applies to different lengths and discontinuous of coating thickness measurement. .

Abstract: In order to overcome the defects of the high computational loads and selecting the threshold of mean square error (MSE) for time domain decision-directed constant modulus blind equalization algorithm (DD+CMA), a frequency domain parallel decision multi-modulus blind equalization algorithm based on frequency domain MMA(FMMA) and frequency domain LMS (FLMS) algorithm is proposed. The proposed algorithm is composed of the FMMA and FLMS, and the FMMA and FLMS run automatically in soft switching parallel manner. In running process, it is not necessary to selecting the threshold of the MSE. Moreover, the computational loads can be reduced by circular convolution in the frequency domain signals instead of linear one of the time domain signals. Simulation results show that performance of the proposed algorithm outperforms the FLMS and the FMMA algorithm.