• We are available for your help 24/7
  • Email: info@isindexing.com, submission@isindexing.com

International Journal of Informatics and Applied Mathematics

Journal Papers (10) Details Call for Paper Manuscript submission Publication Ethics Contact Authors' Guide Line
1 Hybrid Metaheuristic for Optimization Job-Shop Scheduling Problem , Benhamza KARİMA , Zedadra OUARDA 
Real Job-shop scheduling problem is one of the most difficult NP-Combinatorial issues. Exact resolution methods cannot handle large size cases. It is therefore necessary to use heuristic methods to solve them within a reasonable time. There are a large number of metaheuristic, which have the advantage of covering only part of the search space to find an acceptable solution. In this work, Genetic Algorithm and Simulated Annealing are used to solve Job-shop scheduling problem. The objective is to find the sequence of operations on the machines that will minimize the total time required to complete the set of jobs, also known as the "Makespan". Compared to traditional genetic algorithm, hybrid approach yields significant improvement in solution quality.
2 Efficient Camera Clustering Method Based on Overlapping FoVs for WMSNs , Benrazek ALA-EDDİNE, Farou BRAHİM, Kurulay MUHAMMET
The paper presents a new method of clustering cameras in a wireless multimedia sensor network based on overlapping camera fields of view. This method aims to group as many overlapping cameras as possible using the Bron-Kerbosch algorithm. The algorithm allows to find a maximum number of cliques, where they represent camera clusters with strongly overlapping fields of view. The main objective of this method is to form clusters of cameras that have a large overlap area between them in order to restrict the communication area only within the cluster. The method also avoids network congestion and reduce the redundancy of detected data to limit the rapid decrease in energy resulting from the acquisition, processing and transmission of redundant multimedia data. The simulation results show that the proposed method is more effective in extending network lifetime and reducing overhead costs.
3 An Optimized RBF-Neural Network for Breast Cancer Classification , Siouda ROGUİA , Nemissi Mohamed
This paper introduces an optimized RBF-Neural Network for breast cancer classification. The study is based on the optimization of the network through three learning phases. In the first phase, K-means clustering method is used to define RBFs centers. In the second phase, Particle Swarm Optimization is used to optimize RBFs widths. In this phase, a pseudo inverse solution is used to calculate the output weights. Finally, in the third phase, the back-propagation algorithm is used for fine-tuning the obtained parameters, namely centers, widths and output weights. The back-propagation is then initialized with the obtained parameters instead of a random initialization. To evaluate the performance of the proposed method, tests were performed using the Wisconsin Diagnostic Breast Cancer database. The proposed system was compared with a network trained only with BP and a network trained with K-means + PSO. The results obtained are promising compared to other advanced methods and the proposed learning method gives better results by combining these three methods.
4 A New Process for Selecting the Best Background Representatives based on GMM , Nebili WAFA , Seridi Hamid, Kouahla MOHAMED NADJİB
Background subtraction is an essential step in the process of monitoring videos. Several works have been proposed to differentiate the background pixels from the foreground pixels. Mixtures of Gaussian (GMM) are among the most popular models for a such problem. However, they suffer from some inconveniences related to the light variations and complex scene. In this paper, we propose an improvement of the GMM by proposing a new technique of ordering the Gaussian distributions in the selection phase of the best representatives of the scene. Our approach replaces the usual ranking of Gaussians according to the value of wk ,t/σt  with sorting according to their covariance measure which is calculated between each pixel and each of these Gaussians. the obtained results on the Wallflower dataset has proven the effectiveness of the proposed approach compared to standard GMM.
5 Indexing Multimedia Data with an Extension of Binary Tree -- Image Search by Content -- , Kouahla ZİNEDDİNE, Ferrag MOHAMED AMİNE, Anjum ADEEL
Searching for similar images in a data collection, based on a query image, is a fundamental problem for many applications that use large amounts of complex data. Image research by content and on a large scale is a current challenge for large image database research and management. Various information can be extracted such as colour, shape and texture, etc. A characteristic represents only a part of the image property, which makes it necessary to combine all this information to improve the efficiency of these systems. This paper aims to propose a new indexing structure that allows to organize as much information as possible about the images in a binary tree in order to improve the search time, and to propose an algorithm for index construction and a search algorithm for kNN type queries. The concept of containers at the sheet level was used to improve the complexity of algorithms. Experiments on real data sets were conducted to determine its performance.
6 K-means Clustering in R Libraries {cluster} and {factoextra} for Grouping Oceanographic Data , Polina LEMENKOVA
Cluster analysis by k-means algorithm by R programming is the scope of the current paper. The study assesses the similarity of the sampling data derived from the GIS project by homogeneity of their attribute parameters aimed to analyze similar clusters of the observa- tion data by the variety of parameters: geology (similar location on the tectonic plates, sediment thickness, igneous volcanic areas), bathymetry (similar depth ranges) and geomorphology (similar slope steepness and aspect). The geological case study is Mariana Trench. Clustering as ef- fective statistical method to detect similar groups in the data set. Tech- nically, major used R libraries include {cluster}, {factoextra}, {ggplot2}. Minor R libraries include {wordcloud}, {tm}. Several clusters were tested from 2 to 7, optical number is 5. The findings include following computed and visualized results illustrated by 8 figures: 1) correlation matrix show- ing crossing correlations in the combination of factors; 2) comparison of the bi-factors in-between the factors revealed pairwise correlation; 3) pairwise comparative analysis enabled to observe an influence on the variables as bi-factors: in response to the decreasing sediment thickness, slope angles go in parallel; 4) the location of the volcanic igneous ar- eas cause a cyclic repetition of the curve for the slope angles, and those of the volcanic zones have correlation with the slope angle and aspect degree. Findings reveals that four variables affect geomorphology of the trench: slope angle, sediment thickness, aspect degree and volcanic ig- neous areas. The paper includes 7 listings of R programming codes for repeatability of the algorithms in similar research.
7 Security of Smart-Meters against Side-Channel-Attacks (SCA) , İqra MUSTAFA , Adeel ANJUM, Kouahla ZİNEDDİNE
The smart meters become an important node for managing information about electric power system so, smart-meter drags cyber security attention in this regard.  In this paper, the protocol for smart meters named as “privacy preserving billing” is used which provides authentication, non-repudiation and integrity by digital signature scheme and zero-knowledge proof. This protocol ensures secrecy and reliability of end to end communication. However, vulnerability lies in integrated circuits of smart meters that can leak sensitive information and side channel attacks (SCA), derive this information from integrated circuits(IC) while it's operating. The most well-known SCA's against smart-meters are electromagnetic radiations, timing and power analysis attacks. Due to side channel attacks integrated circuit’s physical and electrical effects broadcast information related to secret key and have emerged as a major vulnerability to security applications. SCA does not temper IC security as their non-invasiveness observes device under normal conditions. Hence, our ultimate goal is to make circuit of smart-meter immune against side channel attacks, specifically differential power analysis (DPA) attack is main focus, as it is more aggressive than other SCA’s. For this reason, we present basis for SCA resistance and concept of CMOS library. Secondly, the other concept, we introduces is CMOS-based digital isolation that provides immunity to electrical noise and external fields compared to optocouplers for smart-meters.
8 A Multiple-Place Algorithm for Sustainable Foraging Scenarios , Abderahmane BENKİRAT, Ouarda ZEDADRA
We proposed in this paper a Multi-Place Foraging algorithm called Lévy Walk and Firefly Recruiting Algorithm (LWFR). Unlike, most of the literature works on foraging, our foraging robots forage to maintain the survivability of their nests and collaborate to maintain the survivability of other depots when needed. The Proposed algorithm uses: (1) Lévy Walk to search objects;(2) Firefly algorithm to attract robots in neighborhood. The attraction model inspired by the behavior of Fireflies provides an indirect and costless communication. Numerical simulations show that the proposed algorithm can maintain the survivability of different nests.
9 Physical structure extraction of Algerian baccalaureate transcripts , Abderrahmane KEFALİ, Ahlem OBEİZİ, Chokri FERKOUS
In recent years, Algerian universities have become aware of the interest of electronic archiving and the digitization of archives for a better management of their documents. The development of systems enabling the analysis and understanding of archival documents became an unavoidable need. The present paper follows this trend; it proposes a system for the analysis of the physical structure of Algerian baccalaureate transcripts, stored in the universities archives. The proposed system proceeds in two phases: 1) preprocessing, in which several operations are applied in order to reduce the noise present in the input images. 2) Segmentation; It starts with the elimination of the transcript border. Then, it extracts the text lines and the blocks, based on RLSA algorithm and the projection profiles analysis. After, it proceeds to the classification of the blocks in three: textual block, table, and graphic. Finally, it recovers textual content from textual blocks and tables.
10 Background subtraction based on a Self-Adjusting MoG , Wafa NEBİLİ, Samir HALLACİ, Brahim FAROU
The diversity in background scenes such as, illumination changes, dynamics of the background, camouflage effect, shadow, etc. is a big deal for moving objects detection methods makes it impossible to manage the multimodality of scenes in video surveillance systems. In this paper we present a new method that allows better detection of moving objects. This method combine the robustness of the Artificial Immune Recognition System (AIRS) with respect to the local variations and the power of Gaussian mixtures (MoG) to model changes at the pixel level. The task of the AIRS is to generate several MoG models for each pixel. This models are filtred through two mecanism: the competition for resources and the development of a candidate memory cell. The best model is merged with the exesting MoG according to the Memory cell introduction process. Obtained results on the Wallflower dataset proved the performance of our system compared to other state-of-the-art methods.