The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Based on wavelet and principal component analysis(PCA), an effective shell texture feature of the coscinodiscus extraction method for classification is proposed in this paper.The feature extraction process involves a normalization of the given image with different sizes followed by shift invariant wavelet transform. The shift invariant feature is computed for subband of wavelet coefficients by PCA...
Training a support vector machine (SVM) on a large-scale sample set is a challenging problem. This paper proposes a sample reduction strategy to pretreat training samples which is realized by a two step procedure: instance reduction and attribute reduction, and the classification model of the SVM is also offered. The experimental results show that the proposed reduction algorithm can effectively remove...
In this paper, we propose a real-time forest fire detection algorithm using artificial neural networks based on dynamic characteristics of fire regions segmented from video images. Fire region is obtained from image with the help of threshold values in HSV color space. Area, roundness and contour are computed for fire regions from each 5 continuous frames. The average and mean square deviation of...
Most researches focus on two costs for building cost-sensitive decision trees, such as, misclassification costs, test costs. And the existing literatures always consider the two costs as the same scales, for instance, dollars. However, in real application, it is difficult for us to regard two costs as same scales, for instance, considering misclassification cost as a dollar unit. In this paper, a...
In designing low power switching mode power supply (SMPS), its functions, tech-parameter and cost should be considered at the same time. TOP switch has stronger functions than the discrete components. It is easy and flexible to design SMPS based on TOP Switch and the products always get less cost. SMPS based on UC3842 or UC3843 is compared with the one based on TOP switch in the paper. Then an actual...
The main purpose of this paper is to predict stream-way transition with group method of data handling (GMDH). Therefore, the downstream stream-way transition according to the upstream conditions is forecasted by group method of data handling. Five main factors may affect the stream-way transition include inflow position, inflow angle, slope, flow discharge, and sand content of suspended sediment....
Collaborative filtering recommendation algorithm has proved to be one of the most successful algorithms in recommender systems in recent years. However, traditional centralized collaborative filtering system has suffered from its shortage in scalability as their calculation complexity increases quickly both in time and space when the number of the user and item in the rating database increases. As...
Recommender systems represent personalized services that aim at predicting userspsila interest on information items available in the application domain. Collaborative filtering technique has been proved to be one of the most successful techniques in recommendation systems in recent years. Poor quality is one major challenge in collaborative filtering recommender systems. Sparsity of userspsila ratings...
Aiming at the shortcomings of the BP neural network, this paper presents a method for grain condition information fusion based on BP neural networks and D-S evidential theory. This method firstly employs many BP neural network outputs as the inputs of D-S evidence theory. After that, D-S evidence theory is used to fuse with results from all the neural networks, resulting in the grain quality evaluation...
During last few years, a number of kernel-based online algorithms have been developed that have shown better performance on a number of tasks. A well designed online algorithm needs less computation to reach the same test accuracy as the corresponding batch algorithm. In this paper, we devise an online training algorithm for L2-SVM. Our work is motivated by HULLER, an online algorithm proposed by...
Currently those algorithms to mine the alarm association rules are limited to the minimal support, so that they can only obtain the association rules among the frequently occurring alarms. This paper proposes a new mining algorithm based on spectral graph theory. The algorithms firstly sets up alarm association model with time series; Secondly, it regards alarms database as a high-dimensional structure...
Support vector machine has some advantages, such as simple structure and good generalization, which is one implementation in statistical learning theory. SVM offers a kind of effective way for the data fusion problem of little sample, non-linear and high dimension. In this paper, mobile agents are applied to data fusion system. The model and the study method of data fusion system are improved. An...
There are low bit and high bit hiding algorithms in the spatial domain. The paper proposes a new algorithm. First, divide the cover image into three-pixel blocks. Then, use Arnold transformation to scramble the secret image. Eventually, modify the pixel values to hiding information according to the corresponding mode of size relations and numbers. The experimental results show that the algorithm realizes...
Object recognition in stereo sequences is a simulation of human visual systems on how to analyze and understand various scenes. A pair of stereo sequences is a type of complicated information with huge amount of raw data and features associated with different parameter spaces. Therefore the automatic object recognition in stereo sequences is a difficult and unsolved task challenging many researchers...
On the basis of vibration signal of rolling bearing, a new method of fault diagnosis based on K-L transformation and Lagrange support vector regression is presented.Multidimensional correlated variable is transformed into low dimensional independent eigenvector by the means of K-L transformation. The pattern recognition and nonlinear regression are achieved by the method of Lagrange support vector...
Network tomography techniques can infer the logical topology of network without the cooperation of nodes. Multiple source network tomography can obtain more information about the topology detail and the link performance than single source network tomography. How to differ from each other between the six 2-by-2 structures is the core technique of the multiple source network tomography.This paper proposed...
For reasoning with uncertain knowledge causal semantics analysis is investigated to propose logical rules, which can represent multi-level semantic knowledge of the relationship between the data and information implicated.These rules constitutes several tree structures named decision forest, the number of trees and stopping criteria can be set automatically. Empirical studies on a set of natural domains...
The artificial neural networks (ANNs) were adopted to improve the monitoring capability of water quality in a reservoir using multi-temporal satellite imageries. Simultaneous measurement of chlorophyll-a (Chl-a) concentration along the Feitsui Reservoir, the primary water supply of Taipei City, was conducted by ferryboat. Those ground measured values were used to calibrate empirical functions with...
In order to increase the transmission efficiency of OFDM system, considering the special structure of software radio OFDM system, this paper present a local adaptive least squares support vector regression (LS_SVM) formulation specifically adapted to a pilots-based OFDM signal,and proposes a new method for channel estimation with pilot assisted symbol modulation expectation. The simulation result...
Along with increasing credit cards and growing trade volume in China, credit card fraud rises sharply. How to enhance the detection and prevention of credit card fraud becomes the focus of risk control of banks. This paper proposes a credit card fraud detection model using outlier detection based on distance sum according to the infrequency and unconventionality of fraud in credit card transaction...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.