Retrieval based on skin color to detect the malicious images

Image filtering and retrieval application process has been explored widely. The most basic filtering and retrieval system that is based on extensions of textual database search, is a system where keywords interpreting the images usually describing the certain aspects of the image content. Digital i...

Full description

Saved in:
Bibliographic Details
Main Authors: Nooralisa Mohd Tuah, Nuraini Jamil, Dg. Senandong Ajor
Format: Research Report
Language:English
English
Published: Universiti Malaysia Sabah 2009
Subjects:
Online Access:https://eprints.ums.edu.my/id/eprint/23342/1/Retrieval%20based%20on%20skin%20color%20to%20detect%20the%20malicious%20images.pdf
https://eprints.ums.edu.my/id/eprint/23342/7/retrieval%20based%20on%20color%20to%20detect%20the%20malicious%20images.pdf
https://eprints.ums.edu.my/id/eprint/23342/
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Universiti Malaysia Sabah
Language: English
English
id my.ums.eprints.23342
record_format eprints
spelling my.ums.eprints.233422020-08-26T08:23:22Z https://eprints.ums.edu.my/id/eprint/23342/ Retrieval based on skin color to detect the malicious images Nooralisa Mohd Tuah Nuraini Jamil Dg. Senandong Ajor TA Engineering (General). Civil engineering (General) Image filtering and retrieval application process has been explored widely. The most basic filtering and retrieval system that is based on extensions of textual database search, is a system where keywords interpreting the images usually describing the certain aspects of the image content. Digital image processing algorithms should extract all the relevant features from the image in the same way a human would do. Unfortunately, it is far beyond our knowledge. There still a lot of current algorithms are quite successful in using low-level features of the images such as color histogram and texture. This paper presents an enhancement method of image retrieval. The enhancement method will partitioned the image into sub-regions and the frequency of skin color in each sub-region will be calculated based on skin color and texture features. Over than 80% of skin color is defined as a malicious image. This method was then implemented in our experiment in order to verify whether the enhancement method has successful on determining the image is a malicious image or not. 100 participants have been involved in this experiment where all of them are student from University Malaysia Sabah, Labuan International Campus. Based on the testing and experiment activity, it is clearly showed that our method could retrieve and define a malicious image. However, the limitation is still there to be improved in the future. Universiti Malaysia Sabah 2009 Research Report NonPeerReviewed text en https://eprints.ums.edu.my/id/eprint/23342/1/Retrieval%20based%20on%20skin%20color%20to%20detect%20the%20malicious%20images.pdf text en https://eprints.ums.edu.my/id/eprint/23342/7/retrieval%20based%20on%20color%20to%20detect%20the%20malicious%20images.pdf Nooralisa Mohd Tuah and Nuraini Jamil and Dg. Senandong Ajor (2009) Retrieval based on skin color to detect the malicious images. (Unpublished)
institution Universiti Malaysia Sabah
building UMS Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaysia Sabah
content_source UMS Institutional Repository
url_provider http://eprints.ums.edu.my/
language English
English
topic TA Engineering (General). Civil engineering (General)
spellingShingle TA Engineering (General). Civil engineering (General)
Nooralisa Mohd Tuah
Nuraini Jamil
Dg. Senandong Ajor
Retrieval based on skin color to detect the malicious images
description Image filtering and retrieval application process has been explored widely. The most basic filtering and retrieval system that is based on extensions of textual database search, is a system where keywords interpreting the images usually describing the certain aspects of the image content. Digital image processing algorithms should extract all the relevant features from the image in the same way a human would do. Unfortunately, it is far beyond our knowledge. There still a lot of current algorithms are quite successful in using low-level features of the images such as color histogram and texture. This paper presents an enhancement method of image retrieval. The enhancement method will partitioned the image into sub-regions and the frequency of skin color in each sub-region will be calculated based on skin color and texture features. Over than 80% of skin color is defined as a malicious image. This method was then implemented in our experiment in order to verify whether the enhancement method has successful on determining the image is a malicious image or not. 100 participants have been involved in this experiment where all of them are student from University Malaysia Sabah, Labuan International Campus. Based on the testing and experiment activity, it is clearly showed that our method could retrieve and define a malicious image. However, the limitation is still there to be improved in the future.
format Research Report
author Nooralisa Mohd Tuah
Nuraini Jamil
Dg. Senandong Ajor
author_facet Nooralisa Mohd Tuah
Nuraini Jamil
Dg. Senandong Ajor
author_sort Nooralisa Mohd Tuah
title Retrieval based on skin color to detect the malicious images
title_short Retrieval based on skin color to detect the malicious images
title_full Retrieval based on skin color to detect the malicious images
title_fullStr Retrieval based on skin color to detect the malicious images
title_full_unstemmed Retrieval based on skin color to detect the malicious images
title_sort retrieval based on skin color to detect the malicious images
publisher Universiti Malaysia Sabah
publishDate 2009
url https://eprints.ums.edu.my/id/eprint/23342/1/Retrieval%20based%20on%20skin%20color%20to%20detect%20the%20malicious%20images.pdf
https://eprints.ums.edu.my/id/eprint/23342/7/retrieval%20based%20on%20color%20to%20detect%20the%20malicious%20images.pdf
https://eprints.ums.edu.my/id/eprint/23342/
_version_ 1760230093047726080