Hazem M. Abbas

Automated vision system for localizing structural defects in textile fabrics

Quality control is one of the basic issues in textile industry. Analysis of texture content in digital images plays an important role in the automated visual inspection of textile images to detect their defects. In this paper, a system for automated visual inspection of textiles is discussed. A detailed system configuration is presented and a fault detection algorithm is proposed. Industrial vision systems must operate in real-time, produce a low false alarm rate and be flexible to accommodate variations in inspection sites. This was the rationale behind developing a detection algorithm which employs simple statistical features (mean, variance, median). The intent was to utilize such features to make the calculations simple and fast for the system to be suitable for real-time applications. The performance of the system was evaluated on plain fabrics with different types of textile flaws. The results indicate that the system can detect flaws which vary drastically in physical dimension and nature with a very low false detection rate.

Neural networks for maximum likelihood clustering

Abstract Winner-take-all algorithms are commonly used techniques in clustering analysis. However, they have some problems ranging from clusters under utilization to the extended training time. Some solutions to these problems are addressed here. It is shown here that using the maximum-likelihood criterion instead of the Euclidean distance metric results in better clustering. The clusters are represented by a set of neuron each has a Gaussian receptive field. For these Gaussian neurons, the covariance matrices, in addition to the centers, are learned. The one-winner condition is relaxed by maximizing the likelihood function of the mixture density function of the samples. This produces larger likelihood values and more normally distributed clusters. A fast mixture likelihood clustering is provided for both batch and pattern learning modes. Convergence analysis and experimental results are also presented.

Combinational circuit design using evolutionary algorithms

This paper explores the possibility of enhancing the performance of genetic algorithms (GA) in automating the design of combinational circuits by multi-objective optimization using problem specific genetic operators. The objective is to design digital circuits with 100% functionality and minimum number of logic gates. Experiments are carried out to assess the performance of the proposed algorithm against a conventional genetic algorithm to achieve an error-free circuit (single objective function) with minimum gate representation (multi-objective function).

Enhanced DTW based on-line signature verification

This work describes an enhanced technique for on-line signature verification. The distance between two signatures is computed by dynamic time warping (DTW) method. The reference signatures are used to assign special parameters for each signer, which makes the system cover the intra signer variation. Several features are extracted. Systems with single and multi-features are tested. Curvature change and speed enhance success verification rate. The experiments have been carried out using the SUSIG online signature database. The best result for ROC area under curve is 99.5 with equal error rate 3.48%, and the best result for equal error rate is 3.06% with ROC area under curve 99.43.

Volterra system identification using adaptive genetic algorithms

Abstract In this paper a floating point genetic algorithm (GA) for Volterra system identification is presented. The adaptive GA method suggested here addresses the problem of determining the proper Volterra candidates which leads to the smallest error between the identified nonlinear system and the Volterra model. This is achieved by using variable length GA chromosomes which encode the coefficients of the selected candidates. During the process of evolution the candidates with the least significant contribution in the error reduction process is removed. The proposed GA method detects the proper Volterra candidates and the associated coefficients in one single evolutionary process. The fitness function employed by the algorithm prevents irrelevant candidates from taking part of the final solution. Genetic operators are chosen to suit the floating point representation of the genetic data. As the evolution process improves and the method reaches a near-global solution, a local search is implicitly applied by zooming in the search interval of each gene by adaptively changing the boundaries of those intervals. The proposed algorithms has produced excellent results in modeling different nonlinear systems.

ECG data compression via cubic-splines and scan-along polygonal approximation

Abstract A new electrocardiogram (ECG) data compression method (CUSAPA) is presented in this paper. It applies the scan-along polygonal approximation (SAPA) algorithm on the QRS complex of the ECG waveform and the cubic-splines approximation to the S&−Q segment. This method requires a QRS detector as preprocessor to filter out the QRS complex portions. An attribute grammar is developed to locate the best initial spline knot locations which will represent the S−Q segment. With an overall compression ratio greater than four, the quality of the reconstructed signal is well suited for morphological studies when compared to some other techniques (FFT, FOI and SAPA). The proposed algorithm has shown a significant 50 Hz baseline noise reduction. Extensive computer results obtained with an ECG database have demonstrated the efficiency of the proposed algorithm.

An integer-coded evolutionary approach for mixture maximum likelihood clustering

This paper outlines an algorithm for solving the maximum mixture likelihood clustering problem using an integer-coded genetic algorithm (IGA-ML) where a fixed length chromosome encodes the object-to-cluster assignment. The main advantage of the outlined algorithm (IGA-ML) compared with other known algorithms, such as the k-means technique, is that it can successfully discover the correct number of clusters, in addition to carrying out the partitioning process. The algorithm implements a post-fixing sorting mechanism that drastically reduces the searched solution space by eliminating duplicate solutions that appear after applying the genetic operations. Simulation results show the effectiveness of the algorithm especially with the case of overlapping clusters.

Fabric fault classification using neural trees

In this paper, the problem of textile quality control is addressed. The basic objective is to classify the most important defects in woven fabrics. The algorithm adopted here is composed of three stages. The first stage is a preprocessing phase where defects are detected and localized. Since every detected defect has its different shape and size, all defects are normalized to a predetermined size. In the second stage, a set of features are calculated for each defect using the Haralick et al (1973) spatial features. During the third and last stage, those features are then used to train a competitive neural tree (CNeT) (Behnke and Karayiannis, 1998) designed to learn in a supervised manner the class associated with each set of features. The network can be then used to test and classify new defects. The approach is experimented with a set of images of fault free and defected textiles and output results are analyzed.

Neural Networks Pipeline for Offline Machine Printed Arabic OCR

In the context of Arabic optical characters recognition, Arabic poses more challenges because of its cursive nature. We purpose a system for recognizing a document containing Arabic text, using a pipeline of three neural networks. The first network model predicts the font size of an Arabic word, then the word is normalized to an 18pt font size that will be used to train the next two models. The second model is used to segment a word into characters. The problem of words segmentation in the Arabic language, as in many similar cursive languages, presents a challenge to the OCR systems. This paper presents a multichannel neural network to solve the offline segmentation of machine-printed Arabic documents. The segmented characters are then fed as an input to a convolutional neural network for Arabic characters recognition. The font size prediction model produced a test accuracy of 99.1%. The accuracy of the segmentation model using one font is 98.9%, while four-font model showed 95.5% accuracy. The whole pipeline showed an accuracy of 94.38% on Arabic Transparent font of size 18pt from APTI data set.

Pareto front analog layout placement using Satisfiability Modulo Theories

This paper presents an analog layout placement tool with emphasis on Pareto front generation. In order to handle the exploding number of analog physical constraints, a new approach based on the use of a Satisfiability Modulo Theories (SMT) solver is suggested. SMT is an area concerned with checking the satisfiability of logical formulas over one or more theories. SMT is usually well-tuned to solve specific problems. To our knowledge, this is the first effort to use SMT to tackle analog placement. The proposed tool implicitly generates multiple layouts that fulfill the given constraints. Therefore, it gives the user the option to choose from the feasible solutions through specifying an aspect ratio or by selecting the optimum solution from the Pareto front of the generated shape function. In contrast to most of the existing techniques, as the number of physical constraints increases the SMT solver processing time decreases. The proposed system yielded layouts with a competitive area and run time compared to other techniques.