Human vision perceptual color based semantic image retrieval with relevance feedback

Mario Humberto Mijes Cruz, Mireya Saraí Garciá Vázquez, Alejandro Ramírez Acosta

Research output: Contribution to conferencePaper

Abstract

© 2018 SPIE. Bridging the semantic gap between the low level visual features extracted by computers such as color, texture or shape and high level semantic concepts perceived by humans is the main challenge in the aim of increasing the precision of semantic results into Content-Based Image Retrieval (CBIR). This challenge has been approached with the technique known as Relevance Feedback (RF). The technique of RF can be applied through two methods, biased subspace learning or query movement. The method of query movement is based on Rocchio algorithm. In this paper, we present a new optimization to technique of Relevance Feedback through query movement to develop a CBIR system with better semantic precision. We make a modification to the input images color channels composition in the additive color space (Red, Green, Blue) and perceptual additive color space (Hue, Saturation, Value), through the images representation with human photopic vision behavior, which provides the semantic perception of the colors. With the proposed representation we obtained a more accurate behavior of the Color Histogram (CH), Color Coherence Vector (CCV) and Local Binary Patterns (LBP) descriptors in Rocchio algorithm, thus, a query movement oriented more to the semantics of the user. The optimization performance was measured with a subset of 137 classes with 100 images each one from Caltech256 object database. The results show a significant improvement in the semantic precision in comparison to the P. Mane RF method with prominent features, as well as the performance of CBIR systems without RF using the mentioned descriptors.
Original languageAmerican English
DOIs
StatePublished - 1 Jan 2018
EventProceedings of SPIE - The International Society for Optical Engineering -
Duration: 1 Jan 2018 → …

Conference

ConferenceProceedings of SPIE - The International Society for Optical Engineering
Period1/01/18 → …

Fingerprint

Color vision
Human Vision
Relevance Feedback
semantics
Image retrieval
Image Retrieval
retrieval
Semantics
Color
Feedback
color
Content-based Image Retrieval
Query
Color Space
Descriptors
Color Histogram
Image Representation
optimization
Performance Optimization
Color Image

Cite this

Mijes Cruz, M. H., Garciá Vázquez, M. S., & Ramírez Acosta, A. (2018). Human vision perceptual color based semantic image retrieval with relevance feedback. Paper presented at Proceedings of SPIE - The International Society for Optical Engineering, . https://doi.org/10.1117/12.2320180
Mijes Cruz, Mario Humberto ; Garciá Vázquez, Mireya Saraí ; Ramírez Acosta, Alejandro. / Human vision perceptual color based semantic image retrieval with relevance feedback. Paper presented at Proceedings of SPIE - The International Society for Optical Engineering, .
@conference{6ce550af3fe244b6be1ccf5148855d6c,
title = "Human vision perceptual color based semantic image retrieval with relevance feedback",
abstract = "{\circledC} 2018 SPIE. Bridging the semantic gap between the low level visual features extracted by computers such as color, texture or shape and high level semantic concepts perceived by humans is the main challenge in the aim of increasing the precision of semantic results into Content-Based Image Retrieval (CBIR). This challenge has been approached with the technique known as Relevance Feedback (RF). The technique of RF can be applied through two methods, biased subspace learning or query movement. The method of query movement is based on Rocchio algorithm. In this paper, we present a new optimization to technique of Relevance Feedback through query movement to develop a CBIR system with better semantic precision. We make a modification to the input images color channels composition in the additive color space (Red, Green, Blue) and perceptual additive color space (Hue, Saturation, Value), through the images representation with human photopic vision behavior, which provides the semantic perception of the colors. With the proposed representation we obtained a more accurate behavior of the Color Histogram (CH), Color Coherence Vector (CCV) and Local Binary Patterns (LBP) descriptors in Rocchio algorithm, thus, a query movement oriented more to the semantics of the user. The optimization performance was measured with a subset of 137 classes with 100 images each one from Caltech256 object database. The results show a significant improvement in the semantic precision in comparison to the P. Mane RF method with prominent features, as well as the performance of CBIR systems without RF using the mentioned descriptors.",
author = "{Mijes Cruz}, {Mario Humberto} and {Garci{\'a} V{\'a}zquez}, {Mireya Sara{\'i}} and {Ram{\'i}rez Acosta}, Alejandro",
year = "2018",
month = "1",
day = "1",
doi = "10.1117/12.2320180",
language = "American English",
note = "Proceedings of SPIE - The International Society for Optical Engineering ; Conference date: 01-01-2018",

}

Mijes Cruz, MH, Garciá Vázquez, MS & Ramírez Acosta, A 2018, 'Human vision perceptual color based semantic image retrieval with relevance feedback', Paper presented at Proceedings of SPIE - The International Society for Optical Engineering, 1/01/18. https://doi.org/10.1117/12.2320180

Human vision perceptual color based semantic image retrieval with relevance feedback. / Mijes Cruz, Mario Humberto; Garciá Vázquez, Mireya Saraí; Ramírez Acosta, Alejandro.

2018. Paper presented at Proceedings of SPIE - The International Society for Optical Engineering, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Human vision perceptual color based semantic image retrieval with relevance feedback

AU - Mijes Cruz, Mario Humberto

AU - Garciá Vázquez, Mireya Saraí

AU - Ramírez Acosta, Alejandro

PY - 2018/1/1

Y1 - 2018/1/1

N2 - © 2018 SPIE. Bridging the semantic gap between the low level visual features extracted by computers such as color, texture or shape and high level semantic concepts perceived by humans is the main challenge in the aim of increasing the precision of semantic results into Content-Based Image Retrieval (CBIR). This challenge has been approached with the technique known as Relevance Feedback (RF). The technique of RF can be applied through two methods, biased subspace learning or query movement. The method of query movement is based on Rocchio algorithm. In this paper, we present a new optimization to technique of Relevance Feedback through query movement to develop a CBIR system with better semantic precision. We make a modification to the input images color channels composition in the additive color space (Red, Green, Blue) and perceptual additive color space (Hue, Saturation, Value), through the images representation with human photopic vision behavior, which provides the semantic perception of the colors. With the proposed representation we obtained a more accurate behavior of the Color Histogram (CH), Color Coherence Vector (CCV) and Local Binary Patterns (LBP) descriptors in Rocchio algorithm, thus, a query movement oriented more to the semantics of the user. The optimization performance was measured with a subset of 137 classes with 100 images each one from Caltech256 object database. The results show a significant improvement in the semantic precision in comparison to the P. Mane RF method with prominent features, as well as the performance of CBIR systems without RF using the mentioned descriptors.

AB - © 2018 SPIE. Bridging the semantic gap between the low level visual features extracted by computers such as color, texture or shape and high level semantic concepts perceived by humans is the main challenge in the aim of increasing the precision of semantic results into Content-Based Image Retrieval (CBIR). This challenge has been approached with the technique known as Relevance Feedback (RF). The technique of RF can be applied through two methods, biased subspace learning or query movement. The method of query movement is based on Rocchio algorithm. In this paper, we present a new optimization to technique of Relevance Feedback through query movement to develop a CBIR system with better semantic precision. We make a modification to the input images color channels composition in the additive color space (Red, Green, Blue) and perceptual additive color space (Hue, Saturation, Value), through the images representation with human photopic vision behavior, which provides the semantic perception of the colors. With the proposed representation we obtained a more accurate behavior of the Color Histogram (CH), Color Coherence Vector (CCV) and Local Binary Patterns (LBP) descriptors in Rocchio algorithm, thus, a query movement oriented more to the semantics of the user. The optimization performance was measured with a subset of 137 classes with 100 images each one from Caltech256 object database. The results show a significant improvement in the semantic precision in comparison to the P. Mane RF method with prominent features, as well as the performance of CBIR systems without RF using the mentioned descriptors.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85054668110&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=85054668110&origin=inward

U2 - 10.1117/12.2320180

DO - 10.1117/12.2320180

M3 - Paper

ER -

Mijes Cruz MH, Garciá Vázquez MS, Ramírez Acosta A. Human vision perceptual color based semantic image retrieval with relevance feedback. 2018. Paper presented at Proceedings of SPIE - The International Society for Optical Engineering, . https://doi.org/10.1117/12.2320180