A new approach for interactive semantic image retrieval using the high level semantics

2008 ◽  
Author(s):  
Mohammad Reza Keyvanpour ◽  
Shabnam Asbaghi
Keyword(s):  
Image Retrieval ◽  
New Approach ◽  
Semantic Image Retrieval ◽  
High Level

Research on High-Level Semantic Image Retrieval

2011 ◽  
Vol 268-270 ◽  
pp. 1427-1432
Author(s):  
Chang Yong Ri ◽  
Min Yao
Keyword(s):  
Image Retrieval ◽  
Image Understanding ◽  
Extraction Methods ◽  
Semantic Gap ◽  
Visual Features ◽  
Semantic Features ◽  
Image Description ◽  
Semantic Image Retrieval ◽  
Image Grammar ◽  
High Level

This paper presented the key problems to shorten “semantic gap” between low-level visual features and high-level semantic features to implement high-level semantic image retrieval. First, introduced ontology based semantic image description and semantic extraction methods based on machine learning. Then, illustrated image grammar on the high-level semantic image understanding and retrieval, and-or graph and context based methods of semantic image. Finally, we discussed the development directions and research emphases in this field.

scholarly journals AN OBJECT PROPERTIES FILTER FOR MULTI-MODALITY ONTOLOGY SEMANTIC IMAGE RETRIEVAL

2017 ◽  
Author(s):  
Mohd Suffian Sulaiman ◽  
Sharifalillah Nordin ◽  
Nursuriati Jamil
Keyword(s):  
Image Retrieval ◽  
Semantic Gap ◽  
Retrieval Algorithm ◽  
Semantic Image Retrieval ◽  
Knowledge Based ◽  
Object Properties ◽  
Resource Identification ◽  
Retrieval Mechanism ◽  
Main Components ◽  
High Level

Ontology is a semantic technology that provides the possible approach to bridge the issue on semantic gap in image retrieval between low-level visual features and high-level human semantic. The semantic gap occurs when there is a discrepancy between the information that is extracted from visual data and the text description. In other words, there is a difference between the computational representation in machine and human natural language. In this paper, an ontology has been utilized to reduce the semantic gap by developing a multi-modality ontology image retrieval with the enhancement of a retrieval mechanism by using the object properties filter. To achieve this, a multi-modality ontology semantic image framework was proposed, comprising of four main components which were resource identification, information extraction, knowledge-based construction and retrieval mechanism. A new approach, namely object properties filter is proposed by customizing the semantic image retrieval algorithm and the graphical user interface to facilitate the user to engage with the machine i.e. computers, in order to enhance the retrieval performance. The experiment results showed that the proposed approach delivered better results compared to the approach that did not use the object properties filter based on probability precision measurement.  

Based on the Semantics of the Low-Level Visual Features Image Retrieval

2012 ◽  
Vol 482-484 ◽  
pp. 512-517
Author(s):  
Xian Wen Zeng ◽  
Xue Dong Shen
Keyword(s):  
Image Retrieval ◽  
Input Parameter ◽  
Paper Analysis ◽  
Image Features ◽  
Visual Features ◽  
Color Feature ◽  
Semantic Image Retrieval ◽  
Low Level ◽  
High Level

This paper analysis the reasons that traditional CBIR can’t support based Semantic image retrieval, and gave a kind of method that Using SVM may solute it. Through studying and Classification, combining HSV Color feature as input parameter ,it realized the connection and map between the high-level semantics and low-level image features .Using this method to retrieve can have proved to get higher accuracy.

scholarly journals Molecular Recognition: Perspective and a New Approach

Sensors ◽  
2021 ◽  
Vol 21 (8) ◽  
pp. 2757
Author(s):  
W. Rudolf Seitz ◽  
Casey J. Grenier ◽  
John R. Csoros ◽  
Rongfang Yang ◽  
Tianyu Ren
Keyword(s):  
Molecular Recognition ◽  
Molecularly Imprinted Polymers ◽  
Binding Sites ◽  
Binding Kinetics ◽  
Molecularly Imprinted ◽  
New Approach ◽  
Imprinted Polymers ◽  
High Affinity ◽  
Water Blocking ◽  
High Level

This perspective presents an overview of approaches to the preparation of molecular recognition agents for chemical sensing. These approaches include chemical synthesis, using catalysts from biological systems, partitioning, aptamers, antibodies and molecularly imprinted polymers. The latter three approaches are general in that they can be applied with a large number of analytes, both proteins and smaller molecules like drugs and hormones. Aptamers and antibodies bind analytes rapidly while molecularly imprinted polymers bind much more slowly. Most molecularly imprinted polymers, formed by polymerizing in the presence of a template, contain a high level of covalent crosslinker that causes the polymer to form a separate phase. This results in a material that is rigid with low affinity for analyte and slow binding kinetics. Our approach to templating is to use predominantly or exclusively noncovalent crosslinks. This results in soluble templated polymers that bind analyte rapidly with high affinity. The biggest challenge of this approach is that the chains are tangled when the templated polymer is dissolved in water, blocking access to binding sites.

Sign in / Sign up

Export Citation Format

Share Document