Maher A. Sid-Ahmed

Human face recognition based on multidimensional PCA and extreme learning machine

In this work, a new human face recognition algorithm based on bidirectional two dimensional principal component analysis (B2DPCA) and extreme learning machine (ELM) is introduced. The proposed method is based on curvelet image decomposition of human faces and a subband that exhibits a maximum standard deviation is dimensionally reduced using an improved dimensionality reduction technique. Discriminative feature sets are generated using B2DPCA to ascertain classification accuracy. Other notable contributions of the proposed work include significant improvements in classification rate, up to hundred folds reduction in training time and minimal dependence on the number of prototypes. Extensive experiments are performed using challenging databases and results are compared against state of the art techniques.

A systolic realization for 2-D digital filters

Systolic hardware realizations for two-dimensional finite-impulse-response (FIR) and infinite-impulse-response (IIR) digital filters are presented. The structure permits the 2-D input data to be scanned row-wise and broadcasted one value at a time to various processing elements. Shift registers are used to store the required data needed by the 2-D recursive equation. A processing element can be implemented in VLSI, and as many of these as are required to satisfy the order of the filter can be used to build the total structure. >

Machine recognition of optically captured machine printed arabic text

Abstract This paper discusses the development of algorithms for the machine recognition of optically captured Arabic characters and their isolation from the printed text. Arabic text is formed from a set of connected or individual fonts. Recognition algorithms based only on individual characters need to be supplemented with a separation algorithm. Moment-invariant descriptors are investigated for the purpose of recognition of individual characters. Meanwhile, to recognize connected machine printed characters, an algorithm for separation of individual characters is developed.

An image processing system for locating craniofacial landmarks

A new automatic target recognition algorithm has been developed to extract craniofacial landmarks from lateral skull X-rays (cephalograms). The locations of these landmarks are used by orthodontists in what is referred to as a cephalometric evaluation. The evaluation assists in the diagnosis of anomalies and in the monitoring of treatments. The algorithm is based on gray-scale mathematical morphology. A statistical approach to training was used to overcome subtle differences in skeletal topographies. Decomposition was used to desensitize the algorithm to size differences. A system was trained to locate 20 landmarks. Tests on 40 X-rays showed an 85% recognition rate on average.

A Comparative Study of Power Supply Architectures in Wireless EV Charging Systems

This paper examines two of the primary power supply architectures being predominantly used for wireless electric vehicle (EV) charging, namely the series LC (SLC) resonant and the hybrid series-parallel (LCL) resonant full-bridge inverter topologies. The study of both of these topologies is presented in the context of designing a 3-kW primary-side controlled stationary wireless EV charger with nominal operating parameters of 30-kHz center frequency, a range of coupling in the neighborhood of 0.18-0.26, and a parallel secondary pick-up with partial series coil compensation. A comparison of both architectures is made in terms of their design methodology, physical size, cost, complexity, and efficiency. It is found that the SLC architecture is 2.45% less costly than the LCL topology. On the other hand, it is observed that the LCL architecture achieves almost 10% higher peak efficiency at rated load and minimum coupling. The study also showed that the SLC topology suffers from poor light load efficiency, while the LCL topology maintains very high efficiency over its full range of coupling and loading. The study also revealed that the capacitor voltage stress is significantly higher in the SLC topology. Finally, it is also shown that the control complexity of the SLC architecture is higher than that of the LCL architecture because of its sensitivity to changes in the reflected secondary impedance, which result in loss of constant current source and ZVS operation unless a suitable combination of parameters are modulated by the closed-loop controller.

AN EFFICIENT AUTOMATIC MASS CLASSIFICATION METHOD IN DIGITIZED MAMMOGRAMS USING ARTIFICIAL NEURAL NETWORK

In this paper we present an efficient computer aided mass classification method in digitized mammograms using Artificial Neural Network (ANN), which performs benign-malignant classification on region of interest (ROI) that contains mass. One of the major mammographic characteristics for mass classification is texture. ANN exploits this important factor to classify the mass into benign or malignant. The statistical textural features used in characterizing the masses are mean, standard deviation, entropy, skewness, kurtosis and uniformity. The main aim of the method is to increase the effectiveness and efficiency of the classification process in an objective manner to reduce the numbers of false-positive of malignancies. Three layers artificial neural network (ANN) with seven features was proposed for classifying the marked regions into benign and malignant and 90.91% sensitivity and 83.87% specificity is achieved that is very much promising compare to the radiologists sensitivity 75%.

3D Shape from Focus and Depth Map Computation Using Steerable Filters

The technique utilized to retrieve spatial information from a sequence of images with varying focus plane is termed as shape from focus (SFF). Traditional SFF techniques perform inadequately due to their inability to deal with images that contain high contrast variations between different regions, shadows, defocused points, noise, and oriented edges. A novel technique to compute SFF and depth map is proposed using steerable filters. Steerable filters, designed in quadrature pairs for better control over phase and orientation, have successfully been applied in many image analysis and pattern recognition schemes. Steerable filters represent architecture to synthesize filters of arbitrary orientation using linear combination of basis filters. Such synthesis is used to determine analytically the filter output as a function of orientation. SFF is computed using steerable filters on variety of image sequences. Quantitative and qualitative performance analyses validate enhanced performance of our proposed scheme.

3-D position sensing using a passive monocular vision system

Passive monocular 3-D position sensing is made possible by a new calibration scheme that relates depth to focus blur through a composite lens and aperture model. The calibration technique enables the recovery of absolute 3-D position coordinates from image coordinates and measured focus blur. A geometric model of the cameras position and orientation in space is used to transform the cameras imaging coordinates into world coordinates. The relationship between the world coordinate system and the screen coordinate system which includes the amount of focus blur, is developed by modeling the camera imaging arrangement. The modeling proceeds first through the perspective view from a pinhole camera located anywhere in space. The cameras lens and aperture system is investigated to find the relationship between depth and focus blur. The aspect ratio of the frame image is considered. Position accuracies comparable to those in stereo based vision systems are possible without the need for solving the difficult point of correspondence problem. >

Character scaling by contour method

Advancement in digital-image processing hardware has given the printing industry novel facilities for capturing fonts and has provided new grounds for character scaling, which is an important issue in typesetting and graphical text. An algorithm for digital character scaling by a contour method is developed and implemented. The algorithm is based on scaling the contour of the character through a transformation. Cubic splines are used to interpolate the discrete samples of the contour character. Final results show no jaggies. The algorithm is applied to Arabic fonts and compared to two other algorithms: replication and telescoping template. The superior performance of the contour method is attributed to the cubic spline fitting, which gives better smoothness to the edge of the character. >

A hidden Markov model-based character extraction method

In this paper a hidden Markov model (HMM)-based binarization algorithm is presented. This algorithm performs well for images with nonuniform background. To test the usefullness of the proposed technique some images of composite documents of printed characters were used. These characters were extracted through the proposed binarization algorithms and used in a commercial OCR. A comparative study of various binarization techniques is also presented.