Sharif Digital Repository

In many applications in human society, there is a need for identity recognition of people.Different methods have been developed for this purpose while using the biometrics is one of the major interests. The biometrics measure the unique physiological, anatomical and behavioural characteristics of people. Among them, face is an interesting biometric which have important advantages over other biometrics and face recognition is known as the most common method that people utilize to recognize each other. However, face recognition suffers from factors such as changes in head pose, illumination and face expression which influence the efficiency of recognition methods. The core of many recent... 

In this thesis, change detection in hyperspectral images is investigated. In this process, two hyperspectral images captured from the same scene but in different time instances are given and we intend to detect the occurred changes in the scene. A specific change detection algorithm contains four different steps; namely, preprocessing, selection of the criterion, postprocessing, and decision making. Dimension reduction as a critical process is also deeply investigated to be performed before classification (for decision making purposes.) Two new methods are proposed in the thesis. The proposed MCRD method is designed for dimension reduction and the MPR method is related to the change... 

Feature extraction in subspace domain for face recognition has attracted growing attention in recent years. Face image shown by a long vector usually belongs to a manifold of intrinsically low dimension. Researchers in face recognition field try to extract these manifolds using algebraic and statistical tools. Recently, the use of multilinear algebra and multidimensional data in various stages of feature extraction and recognition is considered. This approach reduces small sample size problem and computational cost by considering the spatial information in the image. Although these successes, the performance of the methods based of this idea in term of recognition rate in the applications... 

Although dimension reduction techniques for text documents can be used for preprocessing of blogs, these techniques will be more effective if they deal with the nature of the blogs properly. In this project we propose a novel algorithm called PostRank using shallow approach to identify theme of the blog or blog representative words in order to reduce the dimensions of blogs. PostRank uses a graph-based syntactic representation of the weblog by taking into account some structural features of weblog. At the first step it models the blog as a complete graph and assumes the theme of the blog as a query applied to a search engine like Google and each post as a search result. It tries to rank the... 

In multi-label classification problems each instance can simultaneously have multiple labels. In these problems, in addition to the complexities of the input feature space we encounter the complexities of output label space. In the multi-label classification problems, there are dependencies between different labels that need to be considered. Since the dimensionality of the label space in real-world applications can be (very) high, most methods which explicitly model these dependencies are ineffective in practice and recently those methods that transform the label space into a latent space have received attention. A class of these methods which uses output space dimension reduction, first... 

In distributed systems, a huge amount of data is dispersed among different nodes; centralization of this data is infeasible due to communication and storage costs. In addition, Databases with high dimensional data objects are becoming more prevalent is many areas. When the dimensionality increases, the volume of the space increases so fast that the available data becomes sparse. This sparsity is problematic from many aspects. In order to obtain a statistically sound and reliable result, the amount of data needed to support the result often grows exponentially with the dimensionality. Also organizing and searching data often relies on detecting areas where objects form groups with similar... 

The main objective of Control Engineering is to lead dynamic behavior of system to an arbitrary state, accomplishment of such an objective requires a model which somehow represents the dynamic behavior of the system, On the other hand an accurate yet fast and efficient model of a complicated model cannot be easily obtained based on the original inputs, states and outputs of the system. This is true both for mechanistic (white box) and empirical (black box) models. In this project, the previous frame work developed for object oriented plant-wide identification has been extended to cover the model reduction based on various established procedures in this area (e.g., statistical and intelligent... 

Linear dimension reduction is one of the common methods in face recognition. But this method is not efficient in cases which borders of different classes are nonlinear. In these cases dimension reduction increases the error of recognition significantly. In the problem of face recognition, there are several factors which make the borders of classes nonlinear including variation in illumination, position and expression of the face. So nonlinear methods has been proposed for face recognition in the presence of nonlinear factors. One of theses nonlinear methods is "Kernel" trick. In the Kernel method data is transferred to the new space with a nonlinear mapping. This mapping should be chosen... 

DNA Microarray is the state-of-the-art technology in analyzing gene expression data. It has made it possible to measure expression levels of thousand of genes simultaneously. Microarray classification has been widely used in effective diagnosis of cancers and some other biological diseases. But the most challenging issue is the intense asymmetry between the dimensionality of genes and tissue samples which can wreck the classification performance. This dissertation will focus on gene selection and reduction solutions and presents a novel classification scheme which uses both gene selection and dimension reduction in its different stages. We have improved one of the recently proposed topology...