RAFIKA M. HARRABI HARRABI - Academia.edu (original) (raw)

Papers by RAFIKA M. HARRABI HARRABI

Research paper thumbnail of Higher order statistics applied to image segmentation

Recent research has shown that image segmentation, has a great importance in many areas and espec... more Recent research has shown that image segmentation, has a great importance in many areas and especially in the industrial application, such as robot animation, mobile localization, etc...¶ Also, in medical imaging, it is used for tumor detection and in radar imaging, it is used for target detection. This paper deals with the problem of texture segmentation using higher order statistics. We propose a novel form of the third order statistics, extend the general concept of the cooccurrence matrix, and define a frequency matrix. First order, second order and third order statistics are analysed and applied on examples related to image segmentation. It is shown that third order statistics provide higher performance and better segmentation results than other methods. The experimental results are handled on twelve Bordatz textures images and then the obtained results are evaluated on using (i) first order statistics using gray level matrix, (ii) second order statistics using co-occurrence matrix and (iii) the third order statistics using frequency matrix. The experimental results demonstrate the importance of using the high order statistic in texture characterisation for image segmentation.

Research paper thumbnail of Color image segmentation using multi-level thresholding approach and data fusion techniques: application in the breast cancer cells images

Eurasip Journal on Image and Video Processing, May 17, 2012

In this article, we present a new color image segmentation method, based on multilevel thresholdi... more In this article, we present a new color image segmentation method, based on multilevel thresholding and data fusion techniques which aim at combining different data sources associated to the same color image in order to increase the information quality and to get a more reliable and accurate segmentation result. The proposed segmentation approach is conceptually different and explores a new strategy. In fact, instead of considering only one image for each application, our technique consists in combining many realizations of the same image, together, in order to increase the information quality and to get an optimal segmented image. For segmentation, we proceed in two steps. In the first step, we begin by identifying the most significant peaks of the histogram. For this purpose, an optimal multi-level thresholding is used based on the two-stage Otsu optimization approach. In the second step, the evidence theory is employed to merge several images represented in different color spaces, in order to get a final reliable and accurate segmentation result. The notion of mass functions, in the Dempster-Shafer (DS) evidence theory, is linked to the Gaussian distribution, and the final segmentation is achieved, on an input image, expressed in different color spaces, by using the DS combination rule and decision. The algorithm is demonstrated through the segmentation of medical color images. The classification accuracy of the proposed method is evaluated and a comparative study versus existing techniques is presented. The experiments were conducted on an extensive set of color images. Satisfactory segmentation results have been obtained showing the effectiveness and superiority of the proposed method.

Research paper thumbnail of Color Image Segmentation by Multilevel Thresholding using a Two Stage Optimization Approach and Fusion

In this paper, we propose a new color image segmentation method based on a multilevel thresholdin... more In this paper, we propose a new color image segmentation method based on a multilevel thresholding algorithm and data fusion techniques. We have revised the Otsu method for selecting optimal threshold values for both unimodal and bimodal distributions, and tested the performance of the new automatic thresholding method called the TSMO (Two-Stage Multi-level Thresholding) on the color images segmentation. This algorithm is iterative and outperforms Otsu's method by greatly reducing the iterations required for computing the between-class variance in an image. For segmentation, we proceed in two steps. In the first step, we begin by identifying the optimal threshold of the tristimuli (R, G and B). In the second step, segmentation results for the three color components are integrated through the fusion rule, in order to get a final reliable and accurate segmentation result. Experimental segmentation results on medical and textured color images demonstrate the value of combing the thresholding technique and fusion rule for color image segmentation. The obtained results show the robustness of the proposed method.

Research paper thumbnail of Iris Recognition Based on Multilevel Thresholding Technique and Modified Fuzzy c-Means Algorithm

Journal on Artificial Intelligence

Research paper thumbnail of FADS: An Intelligent Fatigue and Age Detection System

Mathematics

Nowadays, the use of public transportation is reducing and people prefer to use private transport... more Nowadays, the use of public transportation is reducing and people prefer to use private transport because of its low cost, comfortable ride, and personal preferences. However, personal transport causes numerous real-world road accidents due to the conditions of the drivers’ state such as drowsiness, stress, tiredness, and age during driving. In such cases, driver fatigue detection is mandatory to avoid road accidents and ensure a comfortable journey. To date, several complex systems have been proposed that have problems due to practicing hand feature engineering tools, causing lower performance and high computation. To tackle these issues, we propose an efficient deep learning-assisted intelligent fatigue and age detection system (FADS) to detect and identify different states of the driver. For this purpose, we investigated several neural computing-based methods and selected the most appropriate model considering its feasibility over edge devices for smart surveillance. Next, we dev...

Research paper thumbnail of Intelligent Image Super-Resolution for Vehicle License Plate in Surveillance Applications

Mathematics

Vehicle license plate images are often low resolution and blurry because of the large distance an... more Vehicle license plate images are often low resolution and blurry because of the large distance and relative motion between the vision sensor and vehicle, making license plate identification arduous. The extensive use of expensive, high-quality vision sensors is uneconomical in most cases; thus, images are initially captured and then translated from low resolution to high resolution. For this purpose, several traditional techniques such as bilinear, bicubic, super-resolution convolutional neural network, and super-resolution generative adversarial network (SRGAN) have been developed over time to upgrade low-quality images. However, most studies in this area pertain to the conversion of low-resolution images to super-resolution images, and little attention has been paid to motion de-blurring. This work extends SRGAN by adding an intelligent motion-deblurring method (termed SRGAN-LP), which helps to enhance the image resolution and remove motion blur from the given images. A comprehens...

Research paper thumbnail of Isotropic and anisotropic filtering techniques for image denoising: A comparative study with classification

2012 16th IEEE Mediterranean Electrotechnical Conference, 2012

Denoising image is considered as a very important pretreatment and a basic operation for meaningf... more Denoising image is considered as a very important pretreatment and a basic operation for meaningful analysis of acquired image. In this paper, a comparative study of image denoising techniques is presented. We analyzed the ineffectiveness of isotropic and anisotropic diffusion and extended the work into the regular anisotropic diffusion. Isotropic diffusion is used at locations with low gradient and total variation based diffusion is used along likely edges. These denoising techniques have been applied to textured and satellite images to illustrate the methodology. The PSNR for the test data available is evaluated and the classification accuracy from these denoising techniques is validated. The experimental results demonstrate the superiority of the regular anisotropic diffusion for image denoining.

Research paper thumbnail of Color image segmentation using automatic thresholding techniques

Eighth International Multi-Conference on Systems, Signals & Devices, 2011

In this paper, entropy and between-class variance based thresholding methods for color images seg... more In this paper, entropy and between-class variance based thresholding methods for color images segmentation are studied. The maximization of the between-class variance (MVI)and the entropy (ME) have been used as a criterion functions to determine an optimal threshold to segment images into nearly homogenous regions. Segmentation results from the two methods are validated and the segmentation sensitivity for the test data available is evluated, and a comparative study between these methods in different color spaces is presented. The experimental results demonstrate the superiority of the MVI method for color image segmentation.

Research paper thumbnail of Color image segmentation using a modified Fuzzy C-Means technique and different color spaces: Application in the breast cancer cells images

2014 1st International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), 2014

In this paper, a new color image segmentation method based on modified Fuzzy c-means and data fus... more In this paper, a new color image segmentation method based on modified Fuzzy c-means and data fusion techniques is presented. The proposed segmentation consists in combining many realizations of the same image, to gether, in order to increase the information quality and to get an optimal segmented image. In the first step, the membership degree of each pixel is determined by applying fuzzy c-means clustering to the information coming from the component images to be combined. The idea is to link at the image pixel level, the notion of measurement functions to that of membership functions in fuzzy logic. In the second step, the fuzzy combination theory is employed to merge the component images of the original image, in order to increase the quality of the information and to obtain an optimal segmented image. Segmentation results from the proposed method are validated and classification accuracy for the test date available is evaluated, and then a comparative study versus existing techniques is presented. Experimental segmentation results of color medical and textured images show the effectiveness of the proposed method.

Research paper thumbnail of Face recognition based on statistical features and SVM classifier

Multimedia Tools and Applications, 2022

In this paper, a face recognition method based on statistical features and Support Vector Machine... more In this paper, a face recognition method based on statistical features and Support Vector Machine (SVM) algorithm is proposed. The statistical analysis is used to extract and select the statistical features, whereas, the SVM algorithm is employed to merge and classify the different features in order to increase the quality of the information and to obtain an optimal Human face recognition. Human face recognition results from the proposed method are validated and the True Success Rate (TSR) for the test data available is evaluated, and then a comparative study versus existing techniques is presented. The experimental results with 400 test images of 40 persons demonstrate the superiority of introducing the statistical features in SVM algorithm for human face recognition. In addition, the recognition speed of our method is faster than the classical SVM algorithm and other existing methods. Experimental results show that the algorithm identifies the face images with accuracy of 99.37%.

Research paper thumbnail of Colour image segmentation using the second order statistics and a modified fuzzy C-means technique

Scientific Research and Essays, 2012

This paper presents a new colour image segmentation method based on Fuzzy C-means technique and t... more This paper presents a new colour image segmentation method based on Fuzzy C-means technique and the second order statistics. The importance of combining statistical features extracted from the cooccurrence matrix and the standard Fuzzy C-Means clustering algorithm in the segmentation context is studied in this paper, to obtain a more reliable and accurate segmentation results. In the first phase of segmentation, a characterization degree is employed to identify the most significant statistical features extracted from the co-occurrence matrix. In the second phase, the Fuzzy C-means (FCM) algorithm is used to cluster the statistical feature vectors into homogeneous regions. Segmentation results from the proposed method are validated and a comparative study versus existing techniques is presented. The experimental results on medical and synthetic colour images demonstrate the superiority of introducing the second order statistics in the Fuzzy C-Means algorithm for colour image segmentation.

Research paper thumbnail of Facial Expression Recognition System Based on SVM and HOG Techniques

Facial expression is one of the most commonly used nonverbal means by humans to transmit internal... more Facial expression is one of the most commonly used nonverbal means by humans to transmit internal emotional states and, therefore, it plays a fundamental role in interpersonal interactions. Although there is a wide range of possible facial expressions, psychologists have identified six fundamental ones (happiness, sadness, surprise, anger, fear and disgust) that are universally recognized. Automatic facial expression recognition (FER) is a topic of growing interest mainly due to the rapid spread of assistive technology applications, as human–robot interaction, where a robust emotional awareness is a key point to best accomplish the assistive task. The proposed work aims to design a robust facial expression recognition system (FER). FER system can be divided into three modules, namely facial registration, feature extraction and classification. The objective of this work is the recognition of facial expressions based the Histogram of Oriented Gradients (HOG) and support vector machine...

Research paper thumbnail of Higher order statistics applied to image segmentation

Recent research has shown that image segmentation, has a great importance in many areas and espec... more Recent research has shown that image segmentation, has a great importance in many areas and especially in the industrial application, such as robot animation, mobile localization, etc...¶ Also, in medical imaging, it is used for tumor detection and in radar imaging, it is used for target detection. This paper deals with the problem of texture segmentation using higher order statistics. We propose a novel form of the third order statistics, extend the general concept of the cooccurrence matrix, and define a frequency matrix. First order, second order and third order statistics are analysed and applied on examples related to image segmentation. It is shown that third order statistics provide higher performance and better segmentation results than other methods. The experimental results are handled on twelve Bordatz textures images and then the obtained results are evaluated on using (i) first order statistics using gray level matrix, (ii) second order statistics using co-occurrence matrix and (iii) the third order statistics using frequency matrix. The experimental results demonstrate the importance of using the high order statistic in texture characterisation for image segmentation.

Research paper thumbnail of Color image segmentation using multi-level thresholding approach and data fusion techniques: application in the breast cancer cells images

Eurasip Journal on Image and Video Processing, May 17, 2012

In this article, we present a new color image segmentation method, based on multilevel thresholdi... more In this article, we present a new color image segmentation method, based on multilevel thresholding and data fusion techniques which aim at combining different data sources associated to the same color image in order to increase the information quality and to get a more reliable and accurate segmentation result. The proposed segmentation approach is conceptually different and explores a new strategy. In fact, instead of considering only one image for each application, our technique consists in combining many realizations of the same image, together, in order to increase the information quality and to get an optimal segmented image. For segmentation, we proceed in two steps. In the first step, we begin by identifying the most significant peaks of the histogram. For this purpose, an optimal multi-level thresholding is used based on the two-stage Otsu optimization approach. In the second step, the evidence theory is employed to merge several images represented in different color spaces, in order to get a final reliable and accurate segmentation result. The notion of mass functions, in the Dempster-Shafer (DS) evidence theory, is linked to the Gaussian distribution, and the final segmentation is achieved, on an input image, expressed in different color spaces, by using the DS combination rule and decision. The algorithm is demonstrated through the segmentation of medical color images. The classification accuracy of the proposed method is evaluated and a comparative study versus existing techniques is presented. The experiments were conducted on an extensive set of color images. Satisfactory segmentation results have been obtained showing the effectiveness and superiority of the proposed method.

Research paper thumbnail of Color Image Segmentation by Multilevel Thresholding using a Two Stage Optimization Approach and Fusion

In this paper, we propose a new color image segmentation method based on a multilevel thresholdin... more In this paper, we propose a new color image segmentation method based on a multilevel thresholding algorithm and data fusion techniques. We have revised the Otsu method for selecting optimal threshold values for both unimodal and bimodal distributions, and tested the performance of the new automatic thresholding method called the TSMO (Two-Stage Multi-level Thresholding) on the color images segmentation. This algorithm is iterative and outperforms Otsu's method by greatly reducing the iterations required for computing the between-class variance in an image. For segmentation, we proceed in two steps. In the first step, we begin by identifying the optimal threshold of the tristimuli (R, G and B). In the second step, segmentation results for the three color components are integrated through the fusion rule, in order to get a final reliable and accurate segmentation result. Experimental segmentation results on medical and textured color images demonstrate the value of combing the thresholding technique and fusion rule for color image segmentation. The obtained results show the robustness of the proposed method.

Research paper thumbnail of Iris Recognition Based on Multilevel Thresholding Technique and Modified Fuzzy c-Means Algorithm

Journal on Artificial Intelligence

Research paper thumbnail of FADS: An Intelligent Fatigue and Age Detection System

Mathematics

Nowadays, the use of public transportation is reducing and people prefer to use private transport... more Nowadays, the use of public transportation is reducing and people prefer to use private transport because of its low cost, comfortable ride, and personal preferences. However, personal transport causes numerous real-world road accidents due to the conditions of the drivers’ state such as drowsiness, stress, tiredness, and age during driving. In such cases, driver fatigue detection is mandatory to avoid road accidents and ensure a comfortable journey. To date, several complex systems have been proposed that have problems due to practicing hand feature engineering tools, causing lower performance and high computation. To tackle these issues, we propose an efficient deep learning-assisted intelligent fatigue and age detection system (FADS) to detect and identify different states of the driver. For this purpose, we investigated several neural computing-based methods and selected the most appropriate model considering its feasibility over edge devices for smart surveillance. Next, we dev...

Research paper thumbnail of Intelligent Image Super-Resolution for Vehicle License Plate in Surveillance Applications

Mathematics

Vehicle license plate images are often low resolution and blurry because of the large distance an... more Vehicle license plate images are often low resolution and blurry because of the large distance and relative motion between the vision sensor and vehicle, making license plate identification arduous. The extensive use of expensive, high-quality vision sensors is uneconomical in most cases; thus, images are initially captured and then translated from low resolution to high resolution. For this purpose, several traditional techniques such as bilinear, bicubic, super-resolution convolutional neural network, and super-resolution generative adversarial network (SRGAN) have been developed over time to upgrade low-quality images. However, most studies in this area pertain to the conversion of low-resolution images to super-resolution images, and little attention has been paid to motion de-blurring. This work extends SRGAN by adding an intelligent motion-deblurring method (termed SRGAN-LP), which helps to enhance the image resolution and remove motion blur from the given images. A comprehens...

Research paper thumbnail of Isotropic and anisotropic filtering techniques for image denoising: A comparative study with classification

2012 16th IEEE Mediterranean Electrotechnical Conference, 2012

Denoising image is considered as a very important pretreatment and a basic operation for meaningf... more Denoising image is considered as a very important pretreatment and a basic operation for meaningful analysis of acquired image. In this paper, a comparative study of image denoising techniques is presented. We analyzed the ineffectiveness of isotropic and anisotropic diffusion and extended the work into the regular anisotropic diffusion. Isotropic diffusion is used at locations with low gradient and total variation based diffusion is used along likely edges. These denoising techniques have been applied to textured and satellite images to illustrate the methodology. The PSNR for the test data available is evaluated and the classification accuracy from these denoising techniques is validated. The experimental results demonstrate the superiority of the regular anisotropic diffusion for image denoining.

Research paper thumbnail of Color image segmentation using automatic thresholding techniques

Eighth International Multi-Conference on Systems, Signals & Devices, 2011

In this paper, entropy and between-class variance based thresholding methods for color images seg... more In this paper, entropy and between-class variance based thresholding methods for color images segmentation are studied. The maximization of the between-class variance (MVI)and the entropy (ME) have been used as a criterion functions to determine an optimal threshold to segment images into nearly homogenous regions. Segmentation results from the two methods are validated and the segmentation sensitivity for the test data available is evluated, and a comparative study between these methods in different color spaces is presented. The experimental results demonstrate the superiority of the MVI method for color image segmentation.

Research paper thumbnail of Color image segmentation using a modified Fuzzy C-Means technique and different color spaces: Application in the breast cancer cells images

2014 1st International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), 2014

In this paper, a new color image segmentation method based on modified Fuzzy c-means and data fus... more In this paper, a new color image segmentation method based on modified Fuzzy c-means and data fusion techniques is presented. The proposed segmentation consists in combining many realizations of the same image, to gether, in order to increase the information quality and to get an optimal segmented image. In the first step, the membership degree of each pixel is determined by applying fuzzy c-means clustering to the information coming from the component images to be combined. The idea is to link at the image pixel level, the notion of measurement functions to that of membership functions in fuzzy logic. In the second step, the fuzzy combination theory is employed to merge the component images of the original image, in order to increase the quality of the information and to obtain an optimal segmented image. Segmentation results from the proposed method are validated and classification accuracy for the test date available is evaluated, and then a comparative study versus existing techniques is presented. Experimental segmentation results of color medical and textured images show the effectiveness of the proposed method.

Research paper thumbnail of Face recognition based on statistical features and SVM classifier

Multimedia Tools and Applications, 2022

In this paper, a face recognition method based on statistical features and Support Vector Machine... more In this paper, a face recognition method based on statistical features and Support Vector Machine (SVM) algorithm is proposed. The statistical analysis is used to extract and select the statistical features, whereas, the SVM algorithm is employed to merge and classify the different features in order to increase the quality of the information and to obtain an optimal Human face recognition. Human face recognition results from the proposed method are validated and the True Success Rate (TSR) for the test data available is evaluated, and then a comparative study versus existing techniques is presented. The experimental results with 400 test images of 40 persons demonstrate the superiority of introducing the statistical features in SVM algorithm for human face recognition. In addition, the recognition speed of our method is faster than the classical SVM algorithm and other existing methods. Experimental results show that the algorithm identifies the face images with accuracy of 99.37%.

Research paper thumbnail of Colour image segmentation using the second order statistics and a modified fuzzy C-means technique

Scientific Research and Essays, 2012

This paper presents a new colour image segmentation method based on Fuzzy C-means technique and t... more This paper presents a new colour image segmentation method based on Fuzzy C-means technique and the second order statistics. The importance of combining statistical features extracted from the cooccurrence matrix and the standard Fuzzy C-Means clustering algorithm in the segmentation context is studied in this paper, to obtain a more reliable and accurate segmentation results. In the first phase of segmentation, a characterization degree is employed to identify the most significant statistical features extracted from the co-occurrence matrix. In the second phase, the Fuzzy C-means (FCM) algorithm is used to cluster the statistical feature vectors into homogeneous regions. Segmentation results from the proposed method are validated and a comparative study versus existing techniques is presented. The experimental results on medical and synthetic colour images demonstrate the superiority of introducing the second order statistics in the Fuzzy C-Means algorithm for colour image segmentation.

Research paper thumbnail of Facial Expression Recognition System Based on SVM and HOG Techniques

Facial expression is one of the most commonly used nonverbal means by humans to transmit internal... more Facial expression is one of the most commonly used nonverbal means by humans to transmit internal emotional states and, therefore, it plays a fundamental role in interpersonal interactions. Although there is a wide range of possible facial expressions, psychologists have identified six fundamental ones (happiness, sadness, surprise, anger, fear and disgust) that are universally recognized. Automatic facial expression recognition (FER) is a topic of growing interest mainly due to the rapid spread of assistive technology applications, as human–robot interaction, where a robust emotional awareness is a key point to best accomplish the assistive task. The proposed work aims to design a robust facial expression recognition system (FER). FER system can be divided into three modules, namely facial registration, feature extraction and classification. The objective of this work is the recognition of facial expressions based the Histogram of Oriented Gradients (HOG) and support vector machine...