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.
After studying the disadvantage of BP neural network which has low convergent speed and trap into local minima easily, an idea of designing a new hybrid neural network model. By using Artificial Bee Colony Algorithm (ABC) to expand the updated space of weight and using the fitness functions to decide the better weight. On the basis, make the acquired better value as the weight of BP neural network...
Neural network has self-learning and adaptive ability, and has strong fault tolerance and robustness, so it has a broad applications in the pattern recognition. In this paper, we adopt an improved BP algorithm-Flexible BP algorithm (RPROP) in the image recognition, and we used it to simulate the image recognition in this field of pattern recognition application . The results of the experiments show...
Moments are widely used in pattern recognition, image processing, computer vision and multi resolution analysis. In this paper, we first printout Gaussian-Hermite moments, and propose a new method to extract the character features based on Gaussian-Hermite moments. Following, for training network, the moment features were inputted to BP network as the parameters; so that, a classifier was realized...
In this paper, BP neural network's powerful non-linear fitting ability is applied to establish the non-linear relationship between tunnel surrounding rock stability and its influencing factors. Tunnel surrounding rock stability classification intelligent method based on BP NN is built and is used to surrounding rock stability classification of FuXi tunnel of Tongling-Huangshan highway in Anhui province...
In latest decades credit risk assessment has been a heavy problem in the society especially in the financial system. Credit risk assessment is a decision level decision problem. Information fusion in multi-sensor system is a very complex process, especially in the decision level fusion process. Presently some useful and representative methods, such as neural networks and Dempster-Shafer evidence theory,...
Trained speed of model based on traditional BP neural network was slowly and produced emanative result. A novel land evaluation model based on neural network with genetic optimization algorithm was presented in this paper. The neural network of model is front-network which comprised with five layers architecture which composed of dynamic inference with fuzzy rules where the consequent sub-models are...
This paper researches on the issue of computer recognition to the handwritten character images, including lowercase letters and Arabic numerals. In this paper, we preprocess on characters in order to unified the basic features. And then, we apply the basic method of making the grids to extract the features of character, and classify the respectives. At last, we apply the latest heuristic modifications...
Dissolved gas analysis (DGA) is essential to the fault diagnosis of oil-immersed power transformer. After thoroughly analyzing the gas production mechanism of power transformer faults, it has been found that there are no explicit mapping functions between the single fault of power transformer and the content of gas. To handle this problem, a multi-class classification model for power transformer fault...
Aiming at the low learning rate, bad stability and local minimum problems in standard and some improved BP neural network, in this paper we proposes a novel BP neural network model which concentrates on two aspects: the choice of learning rate and the learning algorithm. In the new model we use Quasic-Newton algorithm to replace gradient descent algorithm or other learning algorithms, thus the new...
Agricultural products information on the Internet is constructed repeatedly, the content is haphazard and sharing resources can not be used, then a classification of improved neural network which is based on the adjustment and optimization of the weight is presented. The adjustment of weight, optimization of network structure and reasonable adjustment of parameters of BP neural network are discussed,...
This paper studies various training algorithms of BP neural network and proposes an improved conjugate gradient algorithm which combines conjugate gradient algorithm with inexact line search route based on generalized Curry principle. The proposed algorithm has global convergence, optimizes the learning steps using new line search rules and improves the convergence speed. The new algorithm is applied...
BP neural network converges slowly and usually falls into partial minimum points. In this paper a new parameter adjustment algorithm is proposed, using anterior accumulated information to modify momentum. The algorithm has characteristics of enhancing the network convergence rate, preventing vibration and reducing network errors. The simulation result indicates that the algorithm is more effective.
The paper is given a new modified differential evolution (MDE) algorithm in which a novel mutation operator is introduced. The MDE algorithm can obtain a good balance between global search and local search and was applied in BP neural network training. The numerical results demonstrate that the new MDE algorithm has the abilities of good global search and faster convergence speed and higher convergence...
In this paper, an on-line signature verification system exploiting local and global information using two-stage fusion is presented. At the first stage, global information is extracted as 13-dimensional vector and recognized by majority classifiers, and then local information is extracted as time functions of various dynamic properties and recognized by BP neural network classifier. By fusing global...
To improve efficiency and quality of case retrieval in case-based reasoning system, a case retrieval model based on the artificial neural network (ANN) and nearest neighbor (NN) algorithm is presented. Firstly, the indexes of cases are created in order to shrink the case-searching range, and the BP neural network is applied to memorize the product cases that are indexed. Secondly, the similar cases,...
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.