OTHERS_CITABLE A Multilevel Thresholding Approach Based on L´evy-Flight Firefly Algorithm for Image Segmentation Multilevel thresholding is an important technique for image processing. The maximum entropy thresholding (MET) has been widely applied in the literature. This paper presented a novel optimal multilevel thresholding approach based on the maximum entropy measure and L´evy-Flight Firefly Algorithm (LFA) for image segmentation. This new method was called, the maximum entropy based on l´evy-flight firefly algorithm for multilevel thresholding (MELFAT) method. In this paper, five famous benchmark images were used to evaluate the proposed method and the results were evaluated by the uniformity measure. The obtained results were compared with five wellknown methods, like Gaussians mooting method (Lim, Y. K., & Lee, S. U. (1990), Symmetry-duality method (Yin, P. Y., & Chen, L. H. (1993), improved GA-based algorithm (Yin, P. -Y. (1999), the hybrid cooperative-comprehensive learning based PSO algorithm (HCOCLPSO) ( Maitra, M., & Chatterjee, A. (2008)) and a new social and momentum component adaptive PSO algorithm (SMCAPSO) (Chander, A.,& Chatterjee, A.,& Siarry, P.(2011)) . The experimental results confirmed the performance and capability of the proposed method to find optimal threshold values. http://ijict.itrc.ac.ir/article-1-190-en.pdf 2018-10-10 1 8 multilevel thresholding entropy image segmentation uniformity levy-flight firefly algorithm Tahereh Hassanzadeh 1 AUTHOR Hakimeh Vojodi 2 AUTHOR Amir Masoud Eftekhari Moghadam 3 AUTHOR
OTHERS_CITABLE A Novel 3D Object Categorization and Retrieval System Using Geometric Features In this paper, we propose a novel geometric features based method to categorize 3D models using probabilistic neural network and support vector machine classifiers. The employed features are extracted from face and vertex characteristics. In addition, we utilize the proposed features in 3D object retrieval. To achieve this end, each model is decomposed into a set of local/global geometrical features. We use histograms of two variables, i.e., deviation angle of normal vector on the object surface point from the vector that connect shape center to that point; and distance of object surface point from shape center. To achieve better separability of different models, mutual Euclidean distance histogram for the pairs of surface points is also used. The most advantage of using histogram to represent the features is that it shows the density of data and enables creating of low dimensional feature vector and consequently decreasing of computational cost in classification process. The effectiveness of our proposed 3D object categorization system has been evaluated on the generalized McGill 3D model dataset in terms of both accuracy and speed measures. Widespread experimental results and comparison with the other similar methods, demonstrate efficiency of the proposed approach to improve both accuracy and speed of categorization system. http://ijict.itrc.ac.ir/article-1-191-en.pdf 2018-10-10 9 20 3D object vertex normal vector center- to-vertex vector mutual Euclidean distance histogram Mohammad Ramezani 1 AUTHOR Hossein Ebrahimnezhad 2 AUTHOR
OTHERS_CITABLE Static Persian Sign Language Recognition Using Kernel-Based Feature Extraction The most effective way for deaf people communication is sign language. Since most people are not familiar with this language, there is a requirement for a sign language translator system. This would be a useful tool specifically in emergency situations. A further need is facilitation of deaf people communication in cyberspace. Sign language gestures can be divided in two groups, including gestures represent the alphabets and those which are arbitrary signs representing specific concepts. The first group is usually introduced by the pose of hands and they are called postures while the second group usually includes motion of the hands. This paper evaluates the efficiency of kernel based feature extraction methods including kernel principle component analysis (KPCA) and kernel discriminant analysis (KDA) on Persian sign language (PSL) postures. To compare the impact of features on signs’ recognition rate, classifiers such as minimum distance (MD), support vector machine (SVM) and Neural network (NN) is used. Experimental trials indicate higher recognition rate for the kernel-based methods in comparison with those of other techniques and also previous works on PSL recognition. http://ijict.itrc.ac.ir/article-1-192-en.pdf 2018-10-10 21 28 Pattern recognition feature extraction kernel-based features support vector machine neural network sign language recognition PSL Milad Moghaddam 1 AUTHOR Manoochehr Nahvi 2 AUTHOR Reza P.R. Hasanzadeh 3 AUTHOR
OTHERS_CITABLE Efficient Object Tracking Using Optimized K-means Segmentation and Radial Basis Function Neural Networks In this paper, an improved method for object tracking is proposed using Radial Basis Function Neural Networks. Optimized k-means color segmentation is employed for detecting an object in first frame. Next the pixelbased color features (R, G, B) from object is used for representing object color and color features from surrounding background is extracted and extended to develop an extended background model. The object and extended background color features are used to train Radial Basis Function Neural Network. The trained RBFNN is employed to detect object in subsequent frames while mean-shift procedure is used to track object location. The performance of the proposed tracker is tested with many video sequences. The proposed tracker is illustrated to be able to track object and successfully resolve the problems caused by the camera movement, rotation, shape deformation and 3D transformation of the target object. The proposed tracker is suitable for real-time object tracking due to its low computational complexity. http://ijict.itrc.ac.ir/article-1-193-en.pdf 2018-10-10 29 39 computer vision object tracking k-means segmentation radial basis function neural networks mean shift Alireza Asvadi 1 AUTHOR MohammadReza Karami 2 AUTHOR Yasser Baleghi 3 AUTHOR
OTHERS_CITABLE An Automatic Contrast Enhancement Technique Using Combination of Histogram-Based Methods This paper proposes the combination of histogram based methods in order to achieve an improved contrast enhancement technique. In the proposed method, at first the histogram is modified in a way that deals with the histogram spikes with less computational complexity. A method is then utilized to preserve the mean brightness of image. In addition, black and white stretching is applied to increase the quality of resulting image. A new method is also introduced so that the stretching parameters can be selected proportional to the intensity distribution of each image. The proposed method is fully automatic, robust against noise, and user-friendly. Experimental results indicate the efficiency of the new technique from both of the objective and subjective viewpoints. http://ijict.itrc.ac.ir/article-1-194-en.pdf 2018-10-10 41 47 contrast enhancement histogram equalization automatic control brightness streching Roghayeh Soleymani 1 AUTHOR Mahrokh Ghani Shayesteh 2 AUTHOR
OTHERS_CITABLE Improvement of Supervised Shape Retrieval by Learning the Manifold Space Manifold learning is the technique that aims for finding a constructive way to embed the data from a highdimensional space into a low-dimensional one based on non-linear approaches. In this paper a supervised manifold learning method for shape recognition is proposed. The approach is based on learning the manifold space for training samples, and maps the test samples to the learned space by a Generalized Regression Neural Network (GRNN). The main goal in this paper is to propose a new feature vector to coincide semantic with Euclidean distances. To accomplish this, the desired topological manifold is learnt by a global distance driven non-linear feature extraction method. The experimental results indicated that the geometrical distances between the samples on the manifold space are more related to their semantic distance. To fuse the results of shape recognition based on contour and region based methods, the final result of shape recognition is based on committee decision in three manifold spaces. The experimental results confirmed the effectiveness and the validity of the proposed method. http://ijict.itrc.ac.ir/article-1-195-en.pdf 2018-10-10 49 56 object recognition shape retrieval shape annotation manifold learning non-linear dimension extraction Mohammad Ali Zare Chahooki 1 AUTHOR Nasrollah Moghadam Charkari 2 AUTHOR
OTHERS_CITABLE Content-Based Image Retrieval for Tourism Application Using Handheld Devices This paper proposed a scenario for using a CBIR (Content-Based Image Retrieval) system in tourism application. Several CBIR algorithms are studied and applied for the proposed scenario. An image database specialized for this application is made to be used for the study purpose. The performance of applied methods were evaluated and compared based on known measures. Among the studied CBIR methods, two algorithms perform better for this application. http://ijict.itrc.ac.ir/article-1-196-en.pdf 2018-10-10 57 64 ANMRR Content-Based Image Retrieval Handheld MPEG-7 Mobile Tourism Zobeir Raisi 1 AUTHOR Farahnaz Mohanna 2 AUTHOR Mehdi Rezaei 3 AUTHOR