Content Based Image Retrieval System

2016 ◽  
pp. 345-362
Author(s):  
Mohd Omar ◽  
Khaleel Ahmad ◽  
M.A. Rizvi
Keyword(s):  
Texture Feature ◽  
Image Features ◽  
Modern World ◽  
Computer Assisted ◽  
Indexing And Retrieval ◽  
Image Retrieval System ◽  
Stored Information ◽  
Automated Methods ◽  
Manual Classification ◽  
Search And Retrieval

In a world of virtualization, where we are having a larger source of images and descriptions available to modern world and based on their requirement it has been utilized from stored information, data center or cloud to larger audience, but at same time rising number of images requires good tools to store the data and retrieve data. Along with this there is a major importance of Quick search and retrieval tools for these growing images to retrieve information quickly and accurately. High demand for automated or computer assisted classification, query and retrieval methods is required to access huge image databases because such method will try to overcome the drawback of higher cost of manual classification and retrieval of relevant image. Scope as researchers to develop automated methods in image features for indexing and retrieval of images related to texture, feature and color is in demand.

scholarly journals Content based Image Retrieval System using Combination of Color and Shape Features, and Siamese Neural Network

2019 ◽  
Vol 9 (2S) ◽  
pp. 71-77
Keyword(s):  
Neural Network ◽  
World Wide ◽  
Retrieval System ◽  
Image Features ◽  
Shape Features ◽  
Retrieval Performance ◽  
Novel Approach ◽  
Learning Technique ◽  
Image Retrieval System ◽  
Similar Images

With an advent of technologya huge collection of digital images is formed as repositories on world wide web (WWW). The task of searching for similar images in the repository is difficult. In this paper, retrieval of similar images from www is demonstrated with the help of combination of image features as color and shape and then using Siamese neural network which is constructed to the requirement as a novel approach. Here, one-shot learning technique is used to test the Siamese Neural Network model for retrieval performance. Various experiments are conducted with both the methods and results obtained are tabulated. The performance of the system is evaluated with precision parameter and which is found to be high.Also, relative study is made with existing works.

scholarly journals Computer-assisted framework for machine-learning–based delineation of GTV regions on datasets of planning CT and PET/CT images

2017 ◽  
Vol 58 (1) ◽  
pp. 123-134 ◽  
Author(s):  
Koujiro Ikushima ◽  
Hidetaka Arimura ◽  
Ze Jin ◽  
Hidetake Yabu-uchi ◽  
Jumpei Kuwazuru ◽  
...  
Keyword(s):  
Machine Learning ◽  
Ground Glass Opacity ◽  
Global Optimum ◽  
Image Features ◽  
Support Vector ◽  
Computer Assisted ◽  
Dice Similarity Coefficient ◽  
Radiation Oncologists ◽  
Tumor Region ◽  
Leave One Out

Abstract We have proposed a computer-assisted framework for machine-learning–based delineation of gross tumor volumes (GTVs) following an optimum contour selection (OCS) method. The key idea of the proposed framework was to feed image features around GTV contours (determined based on the knowledge of radiation oncologists) into a machine-learning classifier during the training step, after which the classifier produces the ‘degree of GTV’ for each voxel in the testing step. Initial GTV regions were extracted using a support vector machine (SVM) that learned the image features inside and outside each tumor region (determined by radiation oncologists). The leave-one-out-by-patient test was employed for training and testing the steps of the proposed framework. The final GTV regions were determined using the OCS method that can be used to select a global optimum object contour based on multiple active delineations with a LSM around the GTV. The efficacy of the proposed framework was evaluated in 14 lung cancer cases [solid: 6, ground-glass opacity (GGO): 4, mixed GGO: 4] using the 3D Dice similarity coefficient (DSC), which denotes the degree of region similarity between the GTVs contoured by radiation oncologists and those determined using the proposed framework. The proposed framework achieved an average DSC of 0.777 for 14 cases, whereas the OCS-based framework produced an average DSC of 0.507. The average DSCs for GGO and mixed GGO were 0.763 and 0.701, respectively, obtained by the proposed framework. The proposed framework can be employed as a tool to assist radiation oncologists in delineating various GTV regions.

scholarly journals Classifications of atmospheric circulation

Geografie ◽  
2016 ◽  
Vol 121 (2) ◽  
pp. 300-323
Author(s):  
Jan Stryhal ◽  
Radan Huth
Keyword(s):  
Atmospheric Circulation ◽  
Synoptic Climatology ◽  
Climate Modelling ◽  
Classification Methods ◽  
The Third ◽  
Circulation Patterns ◽  
Basic Approaches ◽  
Automated Methods ◽  
Manual Classification

The goal of the present article is to provide a brief overview of the development and usage of classifications of atmospheric circulation, particularly classifications of circulation patterns. In the first section, the motivation to conduct research into atmospheric circulation and the role of classifications in this research are discussed. In addition, basic approaches to classification are described. In the second section, manual classification methods are introduced; the focus is on those methods that have been widely used in the Czech literature – the synoptic catalogues of Brádka and Hess-Brezowsky. To our knowledge, such an overview has not been published yet. In the third section, the development of automated methods is described and the most commonly used methods are briefly introduced. We conclude with an overview of one of the fastest developing fields in synoptic climatology – the application of circulation classifications to climate modelling.

Content Based Images Retrieval Based on HSV Color Space and GLCM

2014 ◽  
Vol 644-650 ◽  
pp. 4287-4290
Author(s):  
Ching Hun Su ◽  
Huang Sen Chiu ◽  
Tsai Ming Hsieh
Keyword(s):  
Image Retrieval ◽  
Retrieval System ◽  
Color Space ◽  
Texture Feature ◽  
Hsv Color Space ◽  
Retrieval Scheme ◽  
Image Retrieval System ◽  
Retrieval Problem ◽  
Occurrence Matrix ◽  
Better Than

We propose a practical image retrieval scheme to retrieve images efficiently. We succeed in transferring the image retrieval problem to sequences comparison and subsequently using the color sequences comparison along with the texture feature of Gray Level Co-occurrence matrix to compare the images of database. Thus the computational complexity is decreased obviously. Our results illustrate it has virtues of both the content based image retrieval system and a text based image retrieval system. Experimental results reveal that proposed scheme is better than the conventional methodologies.

Multimodal Analysis in Multimedia Using Symbolic Kernels

2011 ◽  
pp. 842-847
Author(s):  
Hrishikesh B. Aradhye ◽  
Chitra Dorai
Keyword(s):  
Data Storage ◽  
Video Clip ◽  
Multimedia Data ◽  
Communications Technology ◽  
Multimodal Analysis ◽  
Automated Content Analysis ◽  
Indexing And Retrieval ◽  
The Face ◽  
Broadcast Programs ◽  
Search And Retrieval

The rapid adoption of broadband communications technology, coupled with ever-increasing capacity-to-price ratios for data storage, has made multimedia information increasingly more pervasive and accessible for consumers. As a result, the sheer volume of multimedia data available has exploded on the Internet in the past decade in the form of Web casts, broadcast programs, and streaming audio and video. However, indexing, search, and retrieval of this multimedia data is still dependent on manual, text-based tagging (e.g., in the form of a file name of a video clip). However, manual tagging of media content is often bedeviled by an inadequate choice of keywords, incomplete and inconsistent terms used, and the subjective biases of the annotator introduced in his or her descriptions of content adversely affecting accuracy in the search and retrieval phase. Moreover, manual annotation is extremely time-consuming, expensive, and unscalable in the face of ever-growing digital video collections. Therefore, as multimedia get richer in content, become more complex in format and resolution, and grow in volume, the urgency of developing automated content analysis tools for indexing and retrieval of multimedia becomes easily apparent.

Classifying Restatements: An Application of Machine Learning and Textual Analytics

2021 ◽  
Author(s):  
Louise Hayes ◽  
J. Efrim Boritz
Keyword(s):  
Machine Learning ◽  
Audit Quality ◽  
Learning Algorithm ◽  
Reporting Quality ◽  
Financial Statements ◽  
Net Income ◽  
Machine Learning Classification ◽  
Large Samples ◽  
Automated Methods ◽  
Manual Classification

Restatements of audited financial statements are used for evaluating reporting quality and audit quality, and for other evaluative purposes. We constructed a machine learning algorithm to classify restatements by management intent based on the language in restatement announcements. Our machine learning classification is as reliable as other commonly used automated methods such as those based on market reaction, restatement direction, and magnitude. Our method does not require a dictionary of words and is applicable when other automated methods are not, for example, when restatements are announced contemporaneously with financial results and when net income is not restated. For large samples, the use of such a classification algorithm is less tedious and less time-consuming, and more consistent, replicable and scalable than manual classification.

Sign in / Sign up

Export Citation Format

Share Document