Sharif Digital Repository

Feature coding has received great attention in recent years as a building block of many image processing algorithms. In particular, the importance of the locality assumption in coding approaches has been studied in many previous works. We review this assumption and claim that using the similarity of data points to a more global set of anchor points does not necessarily weaken the coding method, as long as the underlying structure of the anchor points is considered. We propose to capture the underlying structure by assuming a random walker over the anchor points. We also show that our method is a fast approximation to the diffusion map kernel. Experiments on various data sets show that with a... 

A rotated general regression neural network is presented as an enhancement to the general regression neural network. A variable kernel estimate for multivariate densities is considered. A coordinate transformation is adopted which circumvent the difficulty of predicting multimodal distribution with large variance differences between modes which is associated with the general regression neural network. The proposed technique trains the network in a way that the variance differences between modes is kept small and in the same order. Further, the technique reduces the number of indispensable training parameters to two parameters and lowers the load of the computation as well as the time for... 

In this letter, a spatially weighted sparse unmixing approach is proposed as a front-end for hyperspectral image classification using a linear SVM. The idea is to partition the pixels of a hyperspectral image into a number of disjoint spatial neighborhoods. Since neighboring pixels are often composed of similar materials, their sparse codes are encouraged to have similar sparsity patterns. This is accomplished by means of a reweighted ℓ1 framework where it is assumed that fractional abundances of neighboring pixels are distributed according to a common Laplacian Scale Mixture (LSM) prior with a shared scale parameter. This shared parameter determines which endmembers contribute to the group... 

Sparse representation is a new approach that has received significant attention for image classification and recognition. This paper presents a PCA-based dictionary building for sparse representation and classification of universal facial expressions. In our method, expressive facials images of each subject are subtracted from a neutral facial image of the same subject. Then the PCA is applied to these difference images to model the variations within each class of facial expressions. The learned principal components are used as the atoms of the dictionary. In the classification step, a given test image is sparsely represented as a linear combination of the principal components of six basic... 

Segmentation of aerial images has been a challenging task in recent years. This paper introduces a simple and efficient method for segmentation and classification of aerial images based on a pixel-level classification. To this end, we use the Gabor texture features in HSV color space as our best experienced features for aerial images segmentation and classification. We test different classifiers including KNN, SVM and a classifier based on sparse representation. Comparison of our proposed method with a sample of segmentation pre-process based classification methods shows that our pixel-wise approach results in higher accuracy results with lower computation time  

Bag of words models for feature extraction have demonstrated top-notch performance in image classification. These representations are usually accompanied by a coding method. Recently, methods that code a descriptor giving regard to its nearby bases have proved efficacious. These methods take into account the nonlinear structure of descriptors, since local similarities are a good approximation of global similarities. However, they confine their usage of the global similarities to nearby bases. In this paper, we propose a coding scheme that brings into focus the manifold structure of descriptors, and devise a method to compute the global similarities of descriptors to the bases. Given a local... 

Voluminous amount of videos in today’s world has made the subject of objective (or semi-objective) classification of videos to be very popular. Among the various descriptors used for video classification, SIFT and LIFT can lead to highly accurate classifiers. But, SIFT descriptor does not consider video motion and LIFT is time-consuming. In this paper, a robust descriptor for semi-supervised classification based on video content is proposed. It holds the benefits of LIFT and SIFT descriptors and overcomes their shortcomings to some extent. For extracting this descriptor, the SIFT descriptor is first used and the motion of the extracted keypoints are then employed to improve the accuracy of... 

Achieving high rates of detection in low rates of embedding is still a challenging problem in many steganalysis systems. The newly proposed steganalysis system based on sparse representation classifier has shown remarkable detection rates in low embedding rate. In this paper, we propose a new steganalysis system based on double sparse representation classifier. We compare our proposed method with other steganalysis systems which use different classifier (including nearest neighbor, support vector machine, ensemble support vector machine and sparse representation). In all of our experiments, input features to the classifier are fixed and the ability of classifier is examined. Also we provide... 

Machine learning algorithms often suffer from good generalization in testing domains especially when the training (source) and test (target) domains do not have similar distributions. To address this problem, several domain adaptation techniques have been proposed to improve the performance of the learning algorithms when they face accuracy degradation caused by the domain shift problem. In this paper, we focus on the non-homogeneous distributed target domains and propose a new latent subdomain discovery model to divide the target domain into subdomains while adapting them. It is expected that applying adaptation on subdomains increase the rate of detection in comparing with the situation... 

Achieving high rates of detection in low rates of embedding is still a challenging problem in many steganalysis systems. The newly proposed steganalysis system based on sparse representation classifier has shown remarkable detection rates in low embedding rate. In this paper, we propose a new steganalysis system based on double sparse representation classifier. We compare our proposed method with other steganalysis systems which use different classifier (including nearest neighbor, support vector machine, ensemble support vector machine and sparse representation). In all of our experiments, input features to the classifier are fixed and the ability of classifier is examined. Also we provide... 

The goal of this paper was to create a new method for analyzing the online real estate flyers based on their property types. We created an algorithm which identifies the buildings and windows from the buildings in order to extract some useful features for classifying the flyers. Our novel approach for building recognition has two main steps: 1- Building Detector 2- Region Growing. Our novel window detection algorithm uses vanishing point to identify nearly the best angle for applying window detection. It transforms the 2D image into 3D and rotates the 3D image around the z-axis and picks the appropriate angle based on the vanishing points. Using these two novel techniques we were be able to... 

Land cover classification is one of the most important applications of POLSAR images. In this paper, a hybrid biogeography-based optimization support vector machine (HBBOSVM) has been introduced to classify POLSAR images of RADARSAT 2 in band C acquired from San Francisco, USA. The main purpose of this classification is to minimize the number of features and maximize classification accuracy. The proposed method consists of three main steps: preprocessing, feature selection and classification. As preprocessing, radiometric calibration, speckle reduction and feature extraction have been performed. In the proposed HBBO, the combination of onlooker bee of artificial bee colony (ABC) and... 

In this paper, a new constructive auto-associative neural network performing nonlinear principal component analysis is presented. The developed constructive neural network maps the data nonlinearly into its principal components and preserves the order of principal components at the same time. The weights of the neural network are trained by a combination of Back Propagation (BP) and Genetic Algorithm (GA) which accelerates the training process by preventing local minima. The performance of the proposed method was evaluated by means of two different experiments that illustrated its efficiency. ©2007 IEEE  

Active Learning Method (ALM) is a powerful fuzzy soft computing tool, developed originally in order to promote an engineering realization of human brain. This algorithm, as a macro-level brain imitation, has been inspired by some behavioral specifications of human brain and active learning ability. ALM is an adaptive recursive fuzzy learning algorithm, in which a complex Multi Input, Multi Output system can be represented as a fuzzy combination of several Single-Input, Single-Output systems. SISO systems as associative layer of algorithm capture partial spatial knowledge of sample data space, and enable a granular knowledge resolution tuning mechanism through the learning process. The... 

This paper presents a structured dictionary-based model for hyperspectral data that incorporates both spectral and contextual characteristics of spectral samples. The idea is to partition the pixels of a hyperspectral image into a number of spatial neighborhoods called contextual groups and to model the pixels inside a group as members of a common subspace. That is, each pixel is represented using a linear combination of a few dictionary elements learned from the data, but since pixels inside a contextual group are often made up of the same materials, their linear combinations are constrained to use common elements from the dictionary. To this end, dictionary learning is carried out with a...