Sharif Digital Repository

In large databases, availability of labeled training data is mostly prohibitive in classification. Semi-supervised algorithms are employed to tackle the lack of labeled training data problem. Video databases are the epitome for such a scenario; that is why semi-supervised learning has found its niche in it. Graph-based methods are a promising platform for semi-supervised video classification. Based on the multiview characteristic of video data, different features have been proposed (such as SIFT, STIP and MFCC) which can be utilized to build a graph. In this paper, we have proposed a new classification method which fuses the results of manifold regularization over different graphs. Our... 

Considering the space of constant-weight sequences as the reference set for every optical orthogonal code (OOC) design algorithm, we propose a classification method that preserves the correlation properties of sequences. First, we introduce the circulant matrix representation of optical orthogonal codes and, based on the spectrum of circulant matrices, we define the spectral classification of the set Sn,w of all (0, 1)-sequences with length n, weight w, and the first chip 1. Then, as a method for spectrally classifying the set Sn,w, we discuss an algebraic structure called multiplicative group action. Using the above multiplicative group action, we define an equivalence relation on Sn,w in... 

Nonlinear dimensionality reduction methods have demonstrated top-notch performance in many pattern recognition and image classification tasks. Despite their popularity, they suffer from highly expensive time and memory requirements, which render them inapplicable to large-scale datasets. To leverage such cases we propose a new method called "Path-Based Isomap". Similar to Isomap, we exploit geodesic paths to find the low-dimensional embedding. However, instead of preserving pairwise geodesic distances, the low-dimensional embedding is computed via a path-mapping algorithm. Due to the much fewer number of paths compared to number of data points, a significant improvement in time and memory... 

Real-world image classification, which aims to determine the semantic class of un-labeled images, is a challenging task. In this paper, we focus on two challenges of image classification and propose a method to address both of them simultaneously. The first challenge is that representing images by heterogeneous features, such as color, shape and texture, helps to provide better classification accuracy. The second challenge comes from dissimilarities in the visual appearance of images from the same class (intra class variance) and similarities between images from different classes (inter class relationship). In addition to these two challenges, we should note that the feature space of... 

Traditional learning algorithms cannot perform well in scenarios where training data (source domain data) that are used to learn the model have a different distribution with test data (target domain data). The domain adaptation that intends to compensate this problem is an important capability for an intelligent agent. This paper presents a domain adaptation method which learns to adapt the data distribution of the source domain to that of the target domain where no labeled data of the target domain is available (and just unlabeled data are available for the target domain). Our method jointly learns a low dimensional representation space and an adaptive classifier. In fact, we try to find a... 

Extremely efficient convolutional neural network architectures are one of the most important requirements for limited-resource devices (such as embedded and mobile devices). The computing power and memory size are two important constraints of these devices. Recently, some architectures have been proposed to overcome these limitations by considering specific hardware-software equipment. In this paper, the lightweight residual densely connected blocks are proposed to guaranty the deep supervision, efficient gradient flow, and feature reuse abilities of convolutional neural network. The proposed method decreases the cost of training and inference processes without using any special... 

Convolution Neural Networks (CNNs), despite being one of the most successful image classification methods, are not robust to most geometric transformations (rotation, isotropic scaling) because of their structural constraints. Recently, scale steerable filters have been proposed to allow scale invariance in CNNs. Although these filters enhance the network performance in scaled image classification tasks, they cannot maintain the scale information across the network. In this paper, this problem is addressed. First, a CNN is built with the usage of scale steerable filters. Then, a scale equivariat network is acquired by adding a feature map to each layer so that the scale-related features are... 

In this study, two methods for solving economic dispatch problems, namely Hopfield neural network and λ iteration method are compared. Three sample of power system with 3, 6 and 20 units have been considered. The time required for CPU, for solving economic dispatch of these two systems has been calculated. It has been shown that for on-line economic dispatch, Hopfield neural network is more efficient and the time required for convergence is considerably smaller compared to classical methods. © 2006 IEEE  

Nondestructive estimation of physicochemical properties, post-harvest physiology, and level of ripeness of fruits is essential to their automated harvesting, sorting, and handling. Recent research efforts have identified machine vision systems as a promising noninvasive nondestructive tool for exploring the relationship between physicochemical and appearance characteristics of fruits at various ripening levels. In this regard, the purpose of the current study is to provide an intelligent algorithm for estimating two physical properties including firmness, and soluble solid content (SSC), three chemical properties viz. starch, acidity, and titratable acidity (TA), as well as detection of the... 

Convolutional neural networks (CNNs) are at the heart of several machine learning applications, while they suffer from computational complexity due to their large number of parameters and operations. Recently, all-optical implementation of the CNNs has achieved many attentions, however, the recently proposed optical architectures for CNNs cannot fully utilize the tremendous capabilities of optical processing, due to the required electro-optical conversions in-between successive layers. To implement an all-optical multi-layer CNN, it is essential to optically implement all required operations, namely convolution, summation of channels' output for each convolutional kernel feeding the... 

Determining the shape, area, volume, and direction of flaws using ultrasonic imaging of metallic pieces, is a method estimating the severity of their defects. Different methods are used to process ultrasound images. Among these methods are spectral analyses, statistical, mathematical and intelligent methods. Within each of these, there are some advantages as well as limitations. Prony algorithm, which has been used as a parametric method for extracting exponential components of a signal, has several applications in signal modeling, system identification and classification. In this paper, after simulating pieces of oil pipeline, digital Wavelet transform has been used to reduce the noise of... 

The existing learning methods usually assume that training data and test data follow the same distribution, while this is not always true. Thus, in many cases the performance of these methods on the test data will be severely degraded. In this paper, we study the problem of unsupervised domain adaptation, where no labeled data in the target domain is available. The proposed method first finds a new representation for both the source and the target domain and then learns a prediction function for the classifier by optimizing an objective function which simultaneously tries to minimize the loss function on the source domain while also maximizes the consistency of manifold (which is based on... 

This study describes the activated carbon (AC) preparation for methane storage. Due to the need for the introduction of a model, correlating the effective preparation parameters with the characteristic parameters of the activated carbon, a model was developed by neural networks. In a previous study [Namvar-Asl M, Soltanieh M, Rashidi A, Irandoukht A. Modeling and preparation of activated carbon for methane storage: (I) modeling of activated carbon characteristics with neural networks and response surface method. Proceedings of CESEP07, Krakow, Poland; 2007.], the model was designed with the MATLAB toolboxes providing the best response for the correlation of the characteristics parameters and... 

The problem of time optimal magnetic attitude control is treated and an open loop solution is first obtained using a variational approach. In order to close the control loop, a neural network with time varying weights is proposed as a feedback optimal controller applicable to the time varying nonlinear system. The good robustness and low real-time computational burden of the proposed neuro-controller makes the controller more useful compared to the other control methods. © 2008 IEEE