Hassan Mathkour

Comparative study of soft computing techniques for mobile robot navigation in an unknown environment

Robot navigation and obstacle avoidance using fuzzy logic controller is presented.Soft computing techniques are used to optimize the performance of fuzzy logic.The automatic tuning was done by using three soft computing techniques: GA, PSO, and NN.The best performance in terms of travelling time and speed is based on GA-Fuzzy.The PSO-Fuzzy and Neuro-Fuzzy methods have better performance in terms of distance travelled. An autonomous mobile robot operating in an unstructured environment must be able to deal with dynamic changes of the environment. Navigation and control of a mobile robot in an unstructured environment are one of the most challenging problems. Fuzzy logic control is a useful tool in the field of navigation of mobile robot. In this research, fuzzy logic controller is optimized by integrating fuzzy logic with other soft computing techniques like genetic algorithm, neural networks, and Particle Swarm Optimization (PSO). Soft computing techniques are used in this work to tune the membership function parameters of fuzzy logic controller to improve the navigation performance. Four methods have been designed and implemented: manually constructed fuzzy logic (M-Fuzzy), fuzzy logic with genetic algorithm (GA-Fuzzy), fuzzy logic with neural network (Neuro-Fuzzy), and fuzzy logic with PSO (PSO-Fuzzy). The performances of these approaches are compared through computer simulations and experiment number of scenarios using Khepera III mobile robot platform. Hybrid fuzzy logic controls with soft computing techniques are found to be most efficient for mobile robot navigation. The GA-Fuzzy technique is found to perform better than the other techniques in most of the test scenarios in terms of travelling time and average speed. The performances of both PSO-Fuzzy and Neuro-Fuzzy are found to be better than the other methods in terms of distance travelled. In terms of bending energy, the PSO-Fuzzy and Neuro-Fuzzy are found to be better in simulation results. Although, the M-Fuzzy is found to be better using real experimental results. Hence, the most important system parameter will dictate which of the four methods to use.

Passive detection of image forgery using DCT and local binary pattern

With the development of easy-to-use and sophisticated image editing software, the alteration of the contents of digital images has become very easy to do and hard to detect. A digital image is a very rich source of information and can capture any event perfectly, but because of this reason, its authenticity is questionable. In this paper, a novel passive image forgery detection method is proposed based on local binary pattern (LBP) and discrete cosine transform (DCT) to detect copy–move and splicing forgeries. First, from the chrominance component of the input image, discriminative localized features are extracted by applying 2D DCT in LBP space. Then, support vector machine is used for detection. Experiments carried out on three image forgery benchmark datasets demonstrate the superiority of the method over recent methods in terms of detection accuracy.

Selection criteria for text mining approaches

Text mining include several techniques like categorization of text, clustering, etc.Text mining techniques can be used to finding useful information from documents.We propose some criteria to evaluate the effectiveness of text mining techniques.These proposed criteria can facilitate the selection of appropriate technique. Text mining techniques include categorization of text, summarization, topic detection, concept extraction, search and retrieval, document clustering, etc. Each of these techniques can be used in finding some non-trivial information from a collection of documents. Text mining can also be employed to detect a documents main topic/theme which is useful in creating taxonomy from the document collection. Areas of applications for text mining include publishing, media, telecommunications, marketing, research, healthcare, medicine, etc. Text mining has also been applied on many applications on the World Wide Web for developing recommendation systems. We propose here a set of criteria to evaluate the effectiveness of text mining techniques in an attempt to facilitate the selection of appropriate technique.

A Novel Approach for Hiding Messages in Images

In this paper, a spiral-based Least Significant Bit (LSB) approach for hiding messages in images is presented. The proposed approach is based on the LSB substitution technique applied on RGB color components of BMP images. The key idea is to divide the image into segments and process them differently. The presented approach considers three algorithms corresponding to three spiral substitution. The proposed algorithms are presented and their performance against attacks and processing time are evaluated. The embedding time was compared for the proposed algorithms as function of the message length up to 20 KB. Numerical experimentation suggests that the embedding time varies linearly with the message length. The assessment of the proposed algorithms against attacks was also evaluated using a chi-square analysis technique exhibiting superior performance

Evaluation of Image Forgery Detection Using Multi-Scale Weber Local Descriptors

Due to the maturing of digital image processing techniques, there are many tools that can forge an image easily without leaving visible traces and lead to the problem of the authentication of digital images. Based on the assumption that forgery alters the texture micro-patterns in a digital image and texture descriptors can be used for modeling this change; we employed two stat-of-the-art local texture descriptors: multi-scale Webers law descriptor (multi-WLD) and multi-scale local binary pattern (multi-LBP) for splicing and copy-move forgery detection. As the tamper traces are not visible to open eyes, so the chrominance components of an image encode these traces and were used for modeling tamper traces with the texture descriptors. To reduce the dimension of the feature space and get rid of redundant features, we employed locally learning based (LLB) algorithm. For identifying an image as authentic or tampered, Support vector machine (SVM) was used. This paper presents the thorough investigation for th...

Optimized Gabor features for mass classification in mammography

Display Omitted Key idea is optimizing Gabor filters such that they respond stronger to features that best discriminate normal and abnormal tissues.Contribution is about a strategy based on PSO and incremental clustering for optimizing a Gabor filter bank for accurate detection.Optimized Gabor filter bank is applied on overlapping blocks of ROIs to collect moment-based features from the magnitudes of Gabor responses. Gabor filter bank has been successfully used for false positive reduction problem and the discrimination of benign and malignant masses in breast cancer detection. However, a generic Gabor filter bank is not adapted to multi-orientation and multi-scale texture micro-patterns present in the regions of interest (ROIs) of mammograms. There are two main optimization concerns: how many filters should be in a Gabor filter band and what should be their parameters. Addressing these issues, this work focuses on finding optimizing Gabor filter banks based on an incremental clustering algorithm and Particle Swarm Optimization (PSO). We employ an SVM with Gaussian kernel as a fitness function for PSO. The effect of optimized Gabor filter bank was evaluated on 1024 ROIs extracted from a Digital Database for Screening Mammography (DDSM) using four performance measures (i.e., accuracy, area under ROC curve, sensitivity and specificity) for the above mentioned mass classification problems. The results show that the proposed method enhances the performance and reduces the computational cost. Moreover, the Wilcoxon signed rank test over the significance level of 0.05 reveals that the performance difference between the optimized Gabor filter bank and non-optimized Gabor filter bank is statistically significant.

An Integrated Approach for Protein Structure Prediction Using Artificial Neural Network

Protein prediction is a fundamental problem in Bioinformatics. Protein structure prediction has vital importance in drug design and biotechnology. Huge amount of biological importance data is being produced and there is great need to transcribe the DNA sequences into amino acid sequences because peptide functions perform important role in body functions of species. Exponential growth of genomic data and complex structure of protein make it challenging to predict its structure. In this paper, we are proposing an integrated approach for the prediction of tri-nucleotide base patterns in DNA strands leading to transcription of peptide regions in genomic sequences. The approach comprise of preprocessing of data, transcription engine and post processing of output. The task has been carried out using series of filters that purify the raw data and assign weights to bases for further feeding to central engine. JOONE (Java Object Oriented Neural network) takes input in the form of segmented data and assign to processes at sigmoid layers. Each layer contains processes and feed forward and back propagation techniques make it possible to predict the sample pattern from genomic sequences of variant sizes.

A multi-agent architecture for adaptive E-learning systems using a blackboard agent

This paper describes a multi-agent architecture for adaptive E-learning system. The proposed architecture is composed of several multi-agent levels and intelligent blackboard as an agent. The blackboard agent provides an easy way for agents to communicate and facilitates the cooperation and coordination among them.

Virtual reality in learning, collaboration and behaviour: content, systems, strategies, context designs

Virtual reality technologies and their applications have received increasing attention in recent years from various perspectives. The thriving numbers behind their adoption and exploitation for different purposes have captured the attention of learning technologies specialists, as well as computer engineering and business researchers. Previously, they have tried to decipher the phenomenon of virtual reality, its relationship to research which has already been conducted and its implications for new research opportunities that affect innovations in teaching. From the well-defined scientific and hardware domain described as information technology, we focus on some of the most revolutionary technologies of our times, namely virtual reality (including all aspects of virtual/augmented/ mixed reality). These technologies are the focus of this special issue which aims to foster a scientific debate for the new era of learning systems and collaboration platforms. The focus of the special issue is the scientific debate for the use of virtual reality as a new channel for facilitating learning and collaboration in various contexts, and the detailed analysis of adoption factors related to behaviour. Analysis of the digital enrichment of the learning and collaboration context will promote a number of strategies for effective learning and collaboration. The current applications of virtual reality in learning and collaboration worldwide present a very interesting picture. Several small-/medium-scale information systems provide a variety of services to all stakeholders, including students, academics, managers and professionals. A key strategic shift in the focus of education is evident, from core-knowledge-oriented education to a collaborative-dynamic mediaenriched evolving paradigm (Lytras et al. 2015). It seems that we are at a crossroads where the traditional classroom-based model of learning has to be critically enriched with technology (Lytras et al. 2014). We believe that one of the key elements of this shift in the coming years will be related to virtual reality applications. These widely accepted virtual reality and modern IT systems demonstrate that a wide range of applications are available and present a viable and robust alternative to traditional monolithic solutions to learning, collaboration, behaviour and education. In parallel, a number of surveys in higher eEducation directly link the satisfactions of students to the use of advanced IT in classrooms. The objective of this special issue is to communicate and disseminate recent higher education, computer engineering and business research, and success stories that demonstrate the power of ICT to improve the user experience with the provision of advanced virtual reality research. We are happy to publish in this special issue a number of top quality research papers and we are grateful to the contributors and the reviewers for their hard work. Eleven research papers contribute to the literature and the body of knowledge of the domain.

A Hierarchical Fuzzy Control Design for Indoor Mobile Robot

This paper presents a motion control for an autonomous robot navigation using fuzzy logic motion control and stereo vision based path-planning module. This requires the capability to maneuver in a complex unknown environment. The mobile robot uses intuitive fuzzy rules and is expected to reach a specific target or follow a prespecified trajectory while moving among unforeseen obstacles. The robots mission depends on the choice of the task. In this paper, behavioral-based control architecture is adopted, and each local navigational task is analyzed in terms of primitive behaviors. Our approach is systematic and original in the sense that some of the fuzzy rules are not triggered in face of critical situations for which the stereo vision camera can intervene to unblock the mobile robot.