Mohammad Reza Alsharif

Image Enhancement Based on Logarithmic Transform Coefficient and Adaptive Histogram Equalization

This paper will present an enhancement technique based upon a new application of contrast limited adaptive histograms on transform domain coefficients called logarithmic transform coefficient adaptive histogram equalization (LTAHE). The method is based on the properties of logarithmic transform domain histogram and contrast limited adaptive histogram equalization. A measure of enhancement based on contrast measure with respect to transform will be used as a tool for evaluating the performance of the proposed enhancement technique and for finding optimal values for variables contained in the enhancement. The algorithms performance will be compared quantitatively to classical histogram equalization using the aforementioned measure of enhancement. Experimental results will be presented to show the performance of the proposed algorithm alongside classical histogram equalization.

Medical image enhancement based on nonlinear technique and logarithmic transform coefficient histogram matching

This paper presents a new method of medical image enhancement that improves the visual quality of digital images as well as images that exhibits dark shadows due to limited dynamic range of imaging. In this paper, non linear image enhancement technique is used in transform domain by the way of transform coefficient histogram matching to enhance image. Processing includes global dynamic range correction and local contrast enhancement which is able to enhance the luminance in the dark shadows keeping the overall tonality consistent with that of the input image. Logarithmic transform histogram matching is used which uses the fact that the relation between stimulus and perception is logarithmic. A measure of enhancement based on the transform is used as a tool for evaluating the performance contrast measure with respect of the proposed enhancement technique. The performance of the algorithm is compared quantitatively to classical histogram equalization using the aforementioned measure of enhancement. A number of experimental results over some x-ray and facial images are presented to show the performance of the proposed algorithm alongside classical histogram equalization.

Automatic Facial Skin Segmentation Based on EM Algorithm under Varying Illumination

Facial skin detection is an important step in facial surgical planning like as many other applications. There are many problems in facial skin detection. One of them is that the image features can be severely corrupted due to illumination, noise, and occlusion, where, shadows can cause numerous strong edges. Hence, in this paper, we present an automatic Expectation-Maximization (EM) algorithm for facial skin color segmentation that uses knowledge of chromatic space and varying illumination conditions to correct and segment frontal and lateral facial color images, simultaneously. The proposed EM algorithm leads to a method that allows for more robust and accurate segmentation of facial images. The initialization of the model parameters is very important in convergence of algorithm. For this purpose, we use a method for robust parameter estimation of Gaussian mixture components. Also, we use an additional class, which includes all pixels not modeled explicitly by Gaussian with small variance, by a uniform probability density, and amending the EM algorithm appropriately, in order to obtain significantly better results. Experimental results on facial color images show that accurate estimates of the Gaussian mixture parameters are computed. Also, other results on images presenting a wide range of variations in lighting conditions, demonstrate the efficiency of the proposed color skin segmentation algorithm compared to conventional EM algorithm.

Fruit defect detection from color images using ACM and MFCM algorithms

Segmentation of an image into its components plays an important role in most of the image processing applications. In this article an important application of image processing in determination of fruit quality is studied, and an automatic algorithm is proposed in order to determine fruits skin color defects. Removing of image background and extraction of fruit shape, exactly, at presence of shadow and complex background is considered as an important preprocessing stage. In proposed algorithm at first, background in image is omitted by using active counter model (ACM) algorithm. Finally, the image is segmented using modified FCM (MFCM) algorithm. Experimental results on fruit color images show that proposed algorithm increases accuracy and speed of fruit skin defect detection, considerably.

An Efficient ICA Based Approach to Multiuser Detection in MIMO OFDM Systems

This paper proposes an ICA-based MIMO-OFDM system which efficiently overcomes problems inherent to ICA by using a precise and robust signal reconstruction method. It exploits the predetermined characteristics introduced to transmitted signals by a convolutional encoder at the transmitter to solve permutation indeterminacy, amplitude scaling ambiguity and phase distortion. Since, the introduced characteristics are only dependent on the convolutional code, despite the previous method, the proposed method is channel independent and robust. Moreover, the method is precise, because the accuracy of the introduced characteristics are fulfilled by an optimized convolutional code. We have compared the performance of the proposed MIMO-OFDM system with joint detection (JD) method which estimates the channels by using two training OFDM blocks. Although the JD method is a training based method, the performance of the proposed blind method is favorably comparable over slowly varying channels, and it dominates JD method over fast varying channels.

An optimum ICA based multiuser data separation for short message service

This paper presents a new algorithm for efficient separation of short messages which are mixed in a multi user short message system. Separation of mixed random binary sequences of data is more difficult than mixed sequences of multivalued signals. The proposed algorithm applies Kullback leibler independent component analysis (ICA) over mixed binary sequences of received data. Normally, the length of binary codes of short messages are less than the required length that makes ICA algorithm sufficiently work. To overcome this problem, a random binary tail is inserted after each user short message at the transmitter side. The inserted tails for different users are acquired in a way to conclude the least correlation between them. The optimum choice of random binary tail not only increase the performance of separation by increasing the data length but also by minimizing the correlation between multiuser data.

A New Image Enhancement Method Based on Nonsubsampled Contourlet Transform

This paper presents a new image enhancement method based on Nonsubsampled Contourlet Transform (NSCT). The contourlet transform is a new extension of the wavelet transform that provides a multi-resolution and multi-direction analysis for two dimension images. The NSCT expansion is composed of basis images oriented at various directions in multiple scales, with flexible aspect ratios. Existing image enhancement methods cannot confine the directional edge information of the image. Given this rich set of basis images, the NSCT transform effectively captures direction edges that are the dominant feature in natural images. Each pixel is enhanced using nonlinear mapping functions depending on the category of the edges. Experimental results ascertain that the proposed method gives better performance of image enhancement than other methods.

An Online Synchronous Brain Wave Signal Pattern Classifier with Parallel Processing Optimization for Embedded System Implementation

Commercial Brain Computer Interface applications are currently expanding due to the success of widespread dis-semination of low cost devices. Reducing the cost of a traditional system requires appropriate resources, such as proper software tools for signal processing and characterization. In this paper, a methodology for classifying a set of attention and meditation brain wave signal patterns is presented by means of unsupervised signal feature clustering with batch Self-Organizing Maps (b-SOM) and supervised classification by Support Vector Machine (SVM). Previous research on this matter did not combine both methods and also required an important amount of computation time. With the use of a small square neuron grid by b-SOM and an RBF kernel SVM, a well delimited classifier was obtained. The recognition rate was 70% after parameter tuning. In terms of optimization, the parallel b-SOM algorithm reduced drastically the computation time, allowing online clustering and classification for full length input data.

Application of full-duplex wireless communication system on echo cancellation

In this paper, we intend to realize the data trading between the mobile station and the base station with only one frequency and in simultaneous process. Until now for transmitting and receiving the data, the mobile station and the base station have been used two different frequencies. This is obvious that by using two different frequencies in transceivers side there would be no interferences in transceiver itself. Therefore, utilization of different frequencies in transceiver was common. However, in these days due to the rapid increase of mobile station users, utilized frequencies between mobile and base stations should be allocated efficiently. Otherwise, it would reduce the speed of data sending in mobile stations. In order to get rid of this problem utilization of the same frequency in transmitter and receiver side is suggested. In this research utilization of the same frequency is considered for the mobile and the base station. However, there are near end echo occurs in system. Meanwhile, the echo canceller is implemented in the suggested system in order to cancel near end signal. In this paper, the considered algorithms of Echo canceller are Least Mean Square, Normalized Mean Square and Recursive Least Square. As a consequence, Normalized Least Mean Square Algorithm performed sufficiently to maintain an adequate low learning curve of normalized mean square error.

Design of Adaptive Friction Control of Small-Scaled Wind Turbine System Considering the Distant Observation

The aim of this research is to reduce down the angular velocity against strong wind during the storm considering the distant observation. Our expectation of this small-scaled wind turbine is to operate continuously even in storm days. However, since the small-scaled wind turbine consists of small blade, it may exceed the limitation of angular velocity in storm condition due to the low inertia moment and consequences of centrifugal force. So it would be broken down the system. Our target is to somehow reduce down the angular velocity using the stall factor control. Stall factor control mainly used in the junction of axis of blade of wind turbines gear wheel. Subsequently, the stall factor is operated when the angular velocity exceeds the limitation. Therefore, adaptive controller is designed and in order to evaluate the angular velocity’s behavior phase space method is introduced. As consequences, the stall factor preserves the stability and angular velocity performs under desired value constantly.