Narrow Results

Training methodology of the Back Propagation Network (BPN) is well documented. One aspect of BPN that requires investigation is whether or not the BPN would get trained for a given training data set and architecture. In this paper the behavior of the BPN is analyzed during its training phase considering convergent and divergent training data sets. Evolution of the weights during the training phase was monitored for the purpose of analysis. The evolution of weights was plotted as return map and was characterized by means of fractal dimension. This fractal dimensional analysis of the weight evolution trajectories is used to provide a new insight to understand the behavior of BPN and dynamics in the evolution of weights.

The Global Positioning System (GPS) is a network of satellites, whose original purpose was to provide accurate navigation, guidance, and time transfer to military users. The past decade has also seen rapid concurrent growth in civilian GPS applications, including farming, mining, surveying, marine, and outdoor recreation. One of the most significant of these civilian applications is commercial aviation. A stand-alone civilian user enjoys an accuracy of 100 meters and 300 nanoseconds, 25 meters and 200 nanoseconds, before and after Selective Availability (SA) was turned off. In some applications, high accuracy is required. In this paper, five Neural Networks (NNs) are proposed for acceptable noise reduction of GPS receivers timing data. The paper uses from an actual data collection for evaluating the performance of the methods. An experimental test setup is designed and implemented for this purpose. The obtained experimental results from a Coarse Acquisition (C/A)-code single-frequency GPS receiver strongly support the potential of methods to give high accurate timing. Quality of the obtained results is very good, so that GPS timing RMS error reduce to less than 120 and 40 nanoseconds, with and without SA.

This paper proposes an implementation scheme of K-class classification problem using systems of multiple neural networks. Usually, a multi-class problem is decomposed into simple sub-problems solved independently using similar single neural networks. For the reason that these sub-problems are not equivalent in their complexity, we propose a system that includes reinforced networks destined to solve complicated parts of the entire problem. Our approach is inspired from principles of the multi-classifiers systems and the labeled classification, which aims to improve performances of the networks trained by the Back-Propagation algorithm. We propose two implementation schemes based on both OAO (one-against-all) and OAA (one-against-one). The proposed models are evaluated using iris and human thigh databases.

Classification of blood cell types can be time consuming and susceptible to error due to the different morphological features of the cells. This paper presents a blood cell identification system that simulates a human visual inspection and identification of the three blood cell types. The proposed system uses global pattern averaging to extract cell features, and a neural network to classify the cell type. Two neural networks are investigated and a comparison between these networks is drawn. Experimental results suggest that the proposed system provides fast, simple and efficient identification which can be used in automating laboratory reporting.

Neural networks differ from traditional approaches to image processing in terms of their ability to adapt to regularities in image structure and to self-organize so as to implement directed transformations. Biomedical ultrasonic images are often degraded in quality by noise and other factors, making enhancement techniques particularly important. This paper describes use of back propagation and competitive learning for enhancement and segmentation of ultrasonic images of the eye. Of particular interest is the extension or these technique to segmentation of three-dimensional data sets, where simple thresholding and gradient operations are not entirely successful.

Image binarization is a task to convert gray-level images into bi-level ones. Its underlying notion can be simply thought of as threshold selection. However, the result of binarization will cause significant influence on the process of image recognition or understanding. In this paper we discuss a new binarization method, named CMB (connectionist model binarization), which uses the connectionist model. In the method a gray-level histogram is input to a multilayer network trained with the back-propagation algorithm to obtain a threshold which gives a visually suitable binarized image. From the experimental results, it was verified that CMB is an effective binarization method in comparison with other methods.

In this paper, a measure of sensitivity is defined to evaluate fault tolerance of neural networks and we show that the sensitivity of a link is closely related to the amount of information passed through it. Based on this assumption, we prove that the distribution of output error caused by s-a-0 (stuck at 0) faults in an MLP network has a Gaussian distribution function. UDBP (Uniformly Distributed Back Propagation) algorithm is then introduced to minimize mean and variance of the output error. Then an MLP neural network trained with UDBP, contributes in an Algorithm-Based Fault Tolerant (ABFT) scheme to protect a nonlinear data process block. A systematic real convolution code guarantees that faults representing errors in the processed data will result in notable nonzero values in syndrome sequence. A majority logic decoder can now easily detect and correct single faults by observing the syndrome sequence. Simulation results demonstrating the error detection and correction behavior against random s-a-0 faults are presented too.

The detection and diagnosis of faults in technical systems are of great practical significance and paramount importance for the safe operation of the plant. The early detection of fault can help avoid system shutdown, breakdown and even catastrophe involving human fatalities and material damage. Since the operator cannot monitor all the variables simultaneously, an automated approach is needed for the real time monitoring and diagnosis of the system. This paper presents the design and development of adaptive neuro-fuzzy inference system (ANFIS) model based for the fault detection of pneumatic valve in cooler water spray system in cement industry. The fault detection model is developed by using two different approaches, namely, ANFIS and feed forward network with back propagation algorithm (BPN). The training and testing data required are developed for the ANFIS model and BPN model that were generated at different operating conditions by operating the pneumatic valve and by creating various faults in real time in a laboratory experimental model. The performance of the developed ANFIS model and back propagation were tested and also compared for a total of 19 faults in pneumatic valve used in cooler water spray system. Obtained results of the ANFIS performed better than BPN.

This paper presents a new method to improve back propagation algorithm from getting stuck with local minima problem and slow convergence speeds which caused by neuron saturation in the hidden layer. In this proposed algorithm, each training pattern has its own activation functions of neurons in the hidden layer that are adjusted by the adaptation of gain parameters together with adaptive momentum and learning rate value during the learning process. The efficiency of the proposed algorithm is compared with the conventional back propagation gradient descent and the current working back propagation gradient descent with adaptive gain by means of simulation on three benchmark problems namely iris, glass and thyroid.

The standard Morse code defines the tone ratio (dash/dot) and the silent ratio (dash-space/dotspace) as 3:1. Since human typing ratio can't keep this ratio precisely and the two ratios —tone ratio and silent ratio—are not equal, the Morse code can't be recognized automatically. The requirement of the standard ratio is difficult to satisfy even for an ordinary person. As for the unstable Morse code typing pattern, the auto-recognition algorithms in the literature are not good enough in applications. The disabled persons usually have difficulty in maintaining a stable typing speeds and typing ratios, we therefore adopted an Expert-Gating neural network model to implement in single chip and recognize online unstable Morse codes. Also, we used another method—a linear back propagation recalling algorithm, to implement in single chip and recognize unstable Morse codes. From three person tests: Test one is a cerebral palsy; Test two is a beginner: Test three is a skilled expert, we have the results: in the experiment of test one, we have 91.15% (use 6 characters average time series as thresholds) and 91.54% (learning 26 characters) online average recognition rate; test two have 95.77% and 96.15%, and test three have 98.46% and 99.23% respectively. As for linear back propagation recalling method online recognized rate, we have the results from test one: 92.31% online average recognition rate; test two: 96.15%; and test three 99.23% respectively. So, we concluded: The Expert-Gating neural network and the linear back propagation recalling algorithm have successfully overcome the difficulty of analyzing a severely online unstable Morse code time series and successfully implement in single chip to recognize online unstable Morse code.

We present a soft computing approach to character recognition from printed documents. For feature extraction, we define a number of fuzzy sets on the Hough transform of character pattern pixels and synthesize additional fuzzy sets by t-norms. The height of each t-norm constitutes a feature element and a set of 'n' such feature elements form an n-dimensional feature vector for the character. A 3n-dimensional vector is then generated from the n-dimensional feature vector by defining three linguistic fuzzy sets, namely, weak, moderate and strong. These 3n-dimensional vectors form a Multilayer Perceptron (MLP) input for training by back propagation. The MLP outputs represent fuzzy sets denoting the belongingness of an input pattern to a number of fuzzy pattern classes. The feature set is chosen by optimizing a Feature Quality Index (FQI) using genetic algorithm. A two-state Markov chain is used to model degraded document images for simulation tests. The system can recognize characters with an accuracy of 98%.

Neural networks differ from traditional approaches to image processing in terms of their ability to adapt to regularities in image structure and to self-organize so as to implement directed transformations. Biomedical ultrasonic images are often degraded in quality by noise and other factors, making enhancement techniques particularly important. This paper describes use of back propagation and competitive learning for enhancement and segmentation of ultrasonic images of the eye. Of particular interest is the extension of these techniques to segmentation of three-dimensional data sets, where simple thresholding and gradient operations are not entirely successful.

Image binarization is a task to convert gray-level images into bi-level ones. Its underlying notion can be simply thought of as threshold selection. However, the result of binarization will cause significant influence on the process of image recognition or understanding. In this paper we discuss a new binarization method, named CMB (connectionist model binarization), which uses the connectionist model. In the method a gray-level histogram is input to a multilayer network trained with the back-propagation algorithm to obtain a threshold which gives a visually suitable binarized image. From the experimental results, it was verified that CMB is an effective binarization method in comparison with other methods.

In this chapter, we shall propose an error back propagation scheme with a periodic chaos neuron model. The present learning model is properly characterized in terms of periodic chaos neurons to involve a chaotic dynamics as well as an external or autonomous control of the periodicity. In practice one may confirm that there exists a close relation between the learning process and the chaotic dynamics, and that the chaotic dynamics promotes the learning speed with a certainly high success rate as seen in the conventional non-chaotic models.