Accepted Papers

  • ECG Signal Classification to Normal and Abnormal Classes Using Histogram-based Features
    Rand Al-Yami1, Turky Alotaiby2, Latifah Aljafar1, Saleh A. Alshebeili3 and Jalila Zouhair1, 1Pince Sultan University, Kingdom of Saudi, 2KACST, Kingdom of Saudi Arabia and 3King Saud University, Kingdom of Saudi Arabia

    This paper presents an inter-patient classification method for classifying multi-leads Electrocardiography (ECG) signals into normal and abnormal classes using histogram-based features. This method is composed of two stages; namely, the preprocessing stage and the classification stage. In the preprocessing stage, a small set of normal and abnormal ECG signals are selected and subsequent histograms are estimated for each class. The discriminative bins in each lead are specified using the Two-sample Kolmogorov-Smirnov test function. Based on these bins, the lead’s importance is estimated. These bins of the most important lead are used as feature indices in the classification stage. Three classifiers; Naïve Bayes (NB), Simple Cart (CART), and Naïve Bayes Multinomial (NBM) methods were used in this study. The proposed approach was evaluated using 104 subjects (52 normal and 52 abnormal) from the Physikalisch-TechnischeBundesanstalt (PTB) dataset and achieved a considerably high sensitivity of 96%, a specificity of 92%, and an accuracy of 94% using the CART classifier system.

  • Distributed Architectures for Specific and Real Applications : Optical Character Recognition (OCR) Application as a Case
    Hamdi Hassen1 and Maher Khemakhem2, 1University of Sfax, Tunisia and 2King Abulaziz University, KSA

    This paper aims at showing some added values of distributed architectures in speeding up and maybe improving the accuracy of specific and real applications through the Arabic Cursive handwriting recognition. Indeed, this field of research is challenging for many important and real-world applications such as document authentication, form processing, postal address identification, bank check recognition, manuscripts recognition, interpretation of historical documents and Islamic manuscripts… Therefore, in the last few decades, researchers and research centers have put an enormous effort into developing various and robust techniques which are unfortunately limited for handwriting recognition of small and medium quantities of documents. This paper attempts to review existing handwriting recognition techniques and the current state of the art in cursive handwriting recognition and presents various hardware solutions based on distributed architectures for large scale (large quantities of documents) Arabic handwritten character recognition system. Our main objective in this paper is to prove that distributed architectures can constitute a corner stone for building efficient large scale Arabic handwriting recognition systems. Experiments were conducted on the Omnivore platform: Grid computing Meta-Scheduling system and P2P Technologies in the Department of Mathematics and Computer Science, University of Marburg, Germany, with a real large scaled dataset from the IFN/ENIT database.

  • Secret Image Transmission Through Mosaic Image
    Shahanaz N and Greeshma R, M Dasan Institute of Technology, India

    A secret image hiding scheme is proposed with new security features. This scheme utilizes the mosaic images, which is created from the secret and target images. A mosaic image is similar to that of the target image. The secret image fragments are hidden in the target image by performing appropriate color transformations. The inverse color transformation is performed for the lossless recovery of secret image. The color transformation is controlled by the proper overflow /underflow methods. The relevant information for recovering the secret image is embedded in the mosaic image by a lossless data hiding with the help of a key. Only with the proper key, the secret image is retrieved from the mosaic image.

  • Automatic Personality Assessment of Handwritten English Script
    Sumita Das and Sonali Nimbhorkar, G. H. Raisoni College of Engineering, India

    Handwriting evidences in criminal cases are typically used to determine the writer’s identity. It is also possible to utilize handwriting to discover the personality traits of an individual, which could prove to be significant information in solving the case. By exploring and analyzing all the components of handwriting and construing them, an outline of the writer's personality traits and temperament could be produced. This paper presents a study of various algorithms at each step in order to design an automated system for assessing personality traits by taking handwriting as input. On the basis of detailed evaluation of each algorithm, methodology has been proposed. The proposed approach can be useful in diverse applications such as criminology, company recruitments, career guidance, personality development, etc.

  • A SSIM Based Method for HEVC Intra Coding
    Caifeng Xu , Tongji University, China

    The latest video coding standard High Efficiency Video Coding (HEVC) achieves about 50% bit-rate reduction compared to H.264/AVC under the same perceptual video quality. The coding unit (CU) is recursively divided into a quad-tree based structure from the largest coding unit (LCU) 64×64 to the smallest coding unit (SCU) 8×8.The CU partition is very time consuming.In this paper a SSIM based fast method for HEVC intra coding is proposed to reduce complexity of coding.Experimental results shows that our proposed method performs xx% encoding time reduction on average with only xx% BD-rate increase for HEVC reference software HM 16.0 under all-intra configuration.

  • Clustering HyperSpectral Data
    Arwa Alturki and Ouiem Bchir, King Saud University, Saudi Arabia

    Hyperspectral imaging succeeded to attract researchers’ interests. A significant number of studies investigated hyperspectral image behaviors through classification and clustering techniques. In this paper, we empirically compared five clustering techniques in different hyperspectral data sets. The considered clustering techniques are K-means, K-medoids, fuzzy C-means, hierarchical, and density-based spatial clustering of applications with noise. Four data sets are used to achieve this purpose which is Botswana, Kennedy space centre, Pavia, and Pavia University. Beside the accuracy, we adopted four more similarity measures: Rand statistics, Jaccard coefficient, Fowlkes-Mallows index, and Hubert index. According to accuracy, we found that fuzzy C-means clustering is doing better on Botswana and Pavia data sets, K-means and K-medoids are giving better results on Kennedy space centre data set, and for Pavia University the hierarchical clustering is better.