Loading...
Search for:
image-retrieval
0.008 seconds
Total 62 records
Reducing Semantic Gap in Content-Based Image Retrieval Systems Using Graph Cuts and Fuzzy Relevance Feedback
,
M.Sc. Thesis
Sharif University of Technology
;
Tabandeh, Mahmoud
(Supervisor)
Abstract
Multimedia retrieval systems are gradually playing a critical role in our everyday life to facilitate interacting with massive amount of personal or professional images, music and video archives. So far, many systems have been proposed among them relevance feedback based content based multimedia (especially image) retrievals has been proved to be more effective. However there is still a problem called semantic gap, in finding proper mapping between low-level features used by CBIR systems and user’s high-level concepts. On the other hand graph cuts have been a great powerful tool for solving many computer vision problems. They benefit from robust optimization algorithm called maximum flow/...
A Study on Image Retrieval Methods
, M.Sc. Thesis Sharif University of Technology ; Razvan, Mohammad-Reza (Supervisor) ; Kamali-Tabrizi, Mostafa (Co-Supervisor)
Abstract
Image retrieval refers to the task of finding images related to a query image within an image set. Due to ever-increasing volumes of data, it has become increasingly necessary to find suitable and efficient methods for searching in massive databases. In this thesis, modern image retrieval techniques developed within the last 15 years have been studied, with an aim to satisfy three primary constraints of efficiency, accuracy, and low memory usage. Our focus has been on content-based retrieval; meaning that instead of using text and other information, we directly utilize image features for analysis and processing. To achieve this, we studied two established techniques, the bag-of-words model,...
A Self-Tag Rectifier Model for Automatic Image Annotation
,
Ph.D. Dissertation
Sharif University of Technology
;
Jamzad, Mansour
(Supervisor)
;
Beigy, Hamid
(Co-Supervisor)
Abstract
Automatic image annotation is an image retrieval mechanism to extract relative semantic tags from visual contents. The number of digital images uploaded in the virtual world is rapidly growing every day. Most of those images are not assigned with proper tags or labels. Although automatic image annotation methods are developed to assign proper tags to images, most of these methods assign some irrelevant tags and also sometimes a few relevant tags are missing. So far, the improvements of accuracy in newly developed automatic image annotation methods have been about one or two percent in F1-score compared to the previous methods. To reach much better performance, we analyzed most of the...
Design a Content-Based Color Image Retrieval Using Attention Driven Saliency Map
, M.Sc. Thesis Sharif University of Technology ; Fatemizadeh, Emadeddin (Supervisor)
Abstract
Content Based Image Retrieval (CBIR) is in fact an image search engine which Operates on image Context . in this thesis (project) the aim was to use the Visual attention of humans in detecting the objects in image. in this ability first a salient image of the most important things in the image would be created And after an initial separation , for the final recognition the other features (details) in the image will be used It’s a while that the use of Visual attention models and saliency maps in designing the interfaces between humans and machines has been considered widely. This fact in the design of CBIR systems has not a good background (satisfying history). In this thesis I have...
Content Based Mammogram Image Retrieval Based on the Multiclass Visual Problem
, M.Sc. Thesis Sharif University of Technology ; Fatemizadeh, Emad (Supervisor)
Abstract
In recent years there has been a great effort to enhance the computer-aided diagnosis systems, Since expertise elicited from past resolved cases plays an important role in medical applications, and images acquired from various cases have a great contribution to diagnosis of the abnormalities, Content based medical image retrieval has become an active research area for many scientists. In this project we proposed a new framework to retrieve visually similar images from a large database, in which visual similarity is regarded as much as the semantic category relevance, we used optimized wavelet transform as the multi-resolution analysis of the images and extracted various statistical SGLDM...
Large scale image search
, M.Sc. Thesis Sharif University of Technology ; Moghaddasi, Reza (Supervisor)
Abstract
Searching for images of the same object or scene in a large number of images is a major problem in computer vision. It has many applications specially in the search engines.For the goal of efficient image search, we need descriptors that are not only discriminative, but also short and need small amount of memory.In this thesis we analyze the image search methods in two categories: The first methods are based on converting the existed descriptors such as gist into a compact binary code. The second methods are based on building short descriptors, especially by some modifications in the framework of bag of features descriptor.Finally we will introduce a novel descriptor "bag of codes" which...
The effect of a two steps searching mechanism Using Feature Vectors Related to Image Class in Improving the Performance of CBIR System
,
M.Sc. Thesis
Sharif University of Technology
;
Jamzad, Mansoor
(Supervisor)
;
Manzuri Shalmani, Mohammad Taghi
(Co-Advisor)
Abstract
Nowadays, retrieval is an inseparable part of user activities and due to growing usage of Content-Based Image Retrieval (CBIR), it has become a hot and challenging research topic specially in the past decade. The most important challenge that retrieval systems (including CBIR systems) are facing is the semantic gap between abstractions in the user’s mind and what is searched. One of the ways of dealing with this challenge is getting more information from the user about what he needs and so decreasing the distance between user’s will and what he gives to search engine as the description of his need. In this research, the class of query image is supposed to be given. For using this...
Automatic Image Annotation by Multi-view Non-negative Matrix Factorization
, Ph.D. Dissertation Sharif University of Technology ; Jamzad, Mansour (Supervisor)
Abstract
Nowadays the number of digital images has largely increased because of progress in internet technology. Management of this volume of data needs an efficient system for browsing, categorizing, and searching the images. The goal of this research is to design a system for automatic annotation of unobserved images for better search in image data bases. Automatic image annotation is a multi-label classification problem with many labels which suggests some words for describing the content of an image. Designing AIA systems faces chanllenges like semantic gap between low level image features and high level human expressions (tags), incompelete tags and imbalance images per tags in the datasets....
Content-based Image Retrieval of Clothing Items with Neural Networks
, M.Sc. Thesis Sharif University of Technology ; Kasaei, Shohreh (Supervisor)
Abstract
One of the trends and important topics in computer vision is content based image retrieval. in this subject, we ask image as query from system, then system will search in pre-processed dataset and finds nearest images to the query and return them as result. in this thesis, our goal is to solve this problem in better way for fashion dataset. current solutions will generate bad results in case of rotation in input query or dataset. last recent years, transformers are generated really good results in NLP, then the ViT reproduced same idea in computer vision and gained comparable results due to CNNs. so, we are going to use vision transformers to solve content-based image retrieval problem with...
Content-Based Image Retrieval Using Relevance Feedback and Semi-Supervised Learning
, M.Sc. Thesis Sharif University of Technology ; Manzuri, Mohammad Taghi (Supervisor)
Abstract
Content-Based Image Retrieval has been an active research area in recent years, due to the vast amount of digital media available via the Internet. In this work we formulate contentbased image retrieval as machine learning problem using the relevance feedback technique and propose a learning algorithm adapted to specific properties of image retrieval. After studying specific properties of image retrieval as a machine learning problem, we propose a Bayesian framework for image retrieval based on one-class learning and test it on different image datasets. The proposed method is a kernel based approach and can also utilize domain knowledge in the form of prior knowledge in constructing a model...
Content Based Image Retrieval Using Segmentation Similarity Measure
,
M.Sc. Thesis
Sharif University of Technology
;
Jamzad, Mansour
(Supervisor)
Abstract
Content Based Image Retrieval (CBIR) is a research area in computer vision. This area comprises of two main steps, low level feature extraction such as color, texture and shape extraction and also similarity measures for comparison of images. The challenge in this system is the existence semantic gap between the low level visual features and the high level image semantics. The aim of research in this field is to reduce this semantic gap. In this study the images are divided into regions using Meanshift method, for color segmentation and then moments of each region as color feature are calculated. Also for extracting texture the images are divided into regions using Jseg method, and then...
Image Recovery from Random and Block Losses
, M.Sc. Thesis Sharif University of Technology ; Marvasti, Farrokh (Supervisor)
Abstract
Digital images degrade during transmittion via noisy channels. The goal of this thesis is to propose new methods for image recovery from random and block losses In the first part of the thesis, various techniques for image recovery from random losses will be reviewd and then a method will be proposed based on the correlation among image pixels in the spatial domain. The method is fast, efficient and robust against Gaussian noise. Also a technique will be developed for quality estimation in the recipient. The second part of the thesis devotes image recovery from block losses. After a brief survey for image inpainting techniques we intoduce the concept of image reconstruction using the...
Use of Fuzzy Type 2 in Image/Video Retrieval
, M.Sc. Thesis Sharif University of Technology ; Ghanbari, Mohammad (Supervisor)
Abstract
In content based image retrieval, low level features are used to find similar image. To do this, many system has been proposed by other people, in which in many of them, combination of features in the same time are used as a step of retrieval to increase accuracy. Feature combination are divied in two category: vector based and weight based which in weight based approach, features can get different weight, based on their importance and role in retrieval accuracy. Each image contain different partition, which some of them like background, base on their lower discrimintivity power, have lower importance. Based on our study, some image have powerfull color features and some of them have...
Video activity analysis based on 3D wavelet statistical properties
, Article 11th International Conference on Advanced Communication Technology, ICACT 2009, Phoenix Park, 15 February 2009 through 18 February 2009 ; Volume 3 , 2009 , Pages 2054-2058 ; 17389445 (ISSN); 9788955191387 (ISBN) ; Ghaemmagham, S ; Khalilain, H ; IEEE Communications Society, IEEE ComSoc; IEEE Region 10 and IEEE Daejeon Section; Korean Institute of Communication Sciences, KICS; lEEK Communications Society, IEEK ComSoc; Korean Institute of Information Scientists and Engineers, KIISE; et al ; Sharif University of Technology
2009
Abstract
A video activity analysis is presented based on 3D wavelet transform. Marginal and joint statistics as well as mutual information estimates are extracted. Marginal histograms are approximated by Generalized Gaussian Density (GGD) functions. The mutual information between coefficients -as a quantitative estimate of joint statistics- decreases when the activity in the video increases. The relationship between kurtosis graphs, extracted from joint distributions and video activity, is deduced. Results show that the type of activity in the video can be figured out from Kurtosis curves. The GGD and the Kullback-Leibler distance (KLD) are used to retrieve and locate 96% of videos properly
Using minimum matching for clustering with balancing constraints
, Article 2009 Second ISECS International Colloquium on Computing, Communication, Control, and Management, CCCM 2009, Sanya, 8 August 2009 through 9 August 2009 ; Volume 1 , 2009 , Pages 225-228 ; 9781424442461 (ISBN) ; Abolhassani, H ; Shirali Shahreza, M. H ; Yangzhou University; Guangdong University of Business Studies; Wuhan Institute of Technology; IEEE SMC TC on Education Technology and Training; IEEE Technology Management Council ; Sharif University of Technology
2009
Abstract
Clustering is a major task in data mining which is used in many applications. However, general clustering is inappropriate for many applications where some constraints should be applied. One category of these constraints is the cluster size constraint. In this paper, we propose a new algorithm for solving the clustering with balancing constraints by using the minimum matching. We compare our algorithm with the method proposed by Banerjee and Ghosh that uses stable matching and show that our algorithm converge to the final solution in fewer iterations. ©2009 IEEE
Using geometrical routing for overlay networking in MMOGs
, Article Multimedia Tools and Applications ; Volume 45, Issue 1-3 , 2009 , Pages 61-81 ; 13807501 (ISSN) ; Pakravan, M. R ; Shirmohammadi, S ; Alavi, M. H ; Sharif University of Technology
2009
Abstract
At a first glance, transmitting update information to a geographic region in the virtual space seems to be an attractive primitive in Massively Multiplayer Online Gaming (MMOG) applications where players are constantly moving and need to send updates to their neighbors who are in the same region of the virtual space. The system would become more scalable if entities did not need to keep track of each other or send messages directly to one another. Rather, an entity could just send a message to a specific region in the virtual space (its area of effect), as opposed to sending packets to specific IP addresses, significantly reducing tracking and routing overhead. Fundamentally speaking, update...
User adaptive clustering for large image databases
, Article Proceedings - International Conference on Pattern Recognition, 23 August 2010 through 26 August 2010, Istanbul ; 2010 , Pages 4271-4274 ; 10514651 (ISSN) ; 9780769541099 (ISBN) ; Jamzad, M ; Rabiee, H. R ; Sharif University of Technology
2010
Abstract
Searching large image databases is a time consuming process when done manually. Current CBIR methods mostly rely on training data in specific domains. When source and domain of images are unknown, unsupervised methods provide better solutions. In this work, we use a hierarchical clustering scheme to group images in an unknown and large image database. In addition, the user should provide the current class assignment of a small number of images as a feedback to the system. The proposed method uses this feedback to guess the number of required clusters, and optimizes the weight vector in an iterative manner. In each step, after modification of the weight vector, the images are reclustered. We...
Unsupervised estimation of conceptual classes for semantic image annotation
, Article 2011 19th Iranian Conference on Electrical Engineering, ICEE 2011, 17 May 2011 through 19 May 2011 ; May , 2011 ; 9789644634284 (ISBN) ; Esmaili, H ; Shirazi, A. A. B ; Sharif University of Technology
2011
Abstract
A probabilistic formulation for semantic image annotation and retrieval is proposed. Annotation and retrieval are posed as classification problems where each class is defined as the group of database images labeled with a common semantic label. It is shown that, by establishing this one-to-one correspondence between semantic labels and semantic classes, a minimum probability of error annotation and retrieval are feasible with algorithms that are 1) conceptually simple and 2) computationally efficient. In this article, a content-based image retrieval and annotation architecture is proposed. Its attitude is decreasing the semantic gap by partitioning the image to its semantic regions and using...
Universal image steganalysis using singular values of DCT coefficients
, Article 2013 10th International ISC Conference on Information Security and Cryptology ; 2013 ; Gaemmaghami, S ; Sharif University of Technology
IEEE Computer Society
2013
Abstract
We propose a blind image steganalysis method based on the Singular Value Decomposition (SVD) of the Discrete Cosine Transform (DCT) coefficients that are revisited in this work. We compute geometric mean, mean of log values, and statistical moments (mean, variance and skewness) of the SVDs of the DCT sub-blocks that are averaged over the whole image to construct a 480-element feature vector for steganalysis. These features are fed to a Support Vector Machine (SVM) classifier to discriminate between stego and cover images. Experimental results show that the proposed method outperforms most powerful steganalyzers when applied to some well-known steganography algorithms
Toward real-time image annotation using marginalized coupled dictionary learning
, Article Journal of Real-Time Image Processing ; Volume 19, Issue 3 , 2022 , Pages 623-638 ; 18618200 (ISSN) ; Hosseini, M. M ; Mohammadi Kashani, M ; Amiri, S. H ; Sharif University of Technology
Springer Science and Business Media Deutschland GmbH
2022
Abstract
In most image retrieval systems, images include various high-level semantics, called tags or annotations. Virtually all the state-of-the-art image annotation methods that handle imbalanced labeling are search-based techniques which are time-consuming. In this paper, a novel coupled dictionary learning approach is proposed to learn a limited number of visual prototypes and their corresponding semantics simultaneously. This approach leads to a real-time image annotation procedure. Another contribution of this paper is that utilizes a marginalized loss function instead of the squared loss function that is inappropriate for image annotation with imbalanced labels. We have employed a marginalized...