An Information-Based Color Feature Representation and Its Application in Detecting Adult Images

For many image classification tasks, color histogram is usually employed as an important ＂signature＂ to describe the color distribution of the image and infer the image content. However, most traditional color histograms cannot achieve satisfactory results in many image classification systems. In or...

Full description

Saved in:
Bibliographic Details
Published inShanghai jiao tong da xue xue bao Vol. 16; no. 4; pp. 395 - 401
Main Author 王士林 刘伟聪
Format Journal Article
LanguageEnglish
Published Heidelberg Shanghai Jiaotong University Press 01.08.2011
Subjects
Adults
Architecture
Classification
Color
Computer Science
Electrical Engineering
Engineering
Histograms
Image classification
Life Sciences
Materials Science
Representations
Tasks
mutual information
image classification
color histogram
TP 391.41
Online AccessGet full text
ISSN1007-1172
1995-8188
DOI10.1007/s12204-011-1167-7

Cover

More Information
Summary:For many image classification tasks, color histogram is usually employed as an important ＂signature＂ to describe the color distribution of the image and infer the image content. However, most traditional color histograms cannot achieve satisfactory results in many image classification systems. In order to improve the accuracy and reduce the computational complexity of the classification task, an information-based color feature representation is proposed in this paper. The mutual information between the feature and the class label is adopted to evaluate the discriminative power of the feature. A novel quantization scheme is presented, which removes the redundant color components and combines the adjacent components together to generate a new feature to maximize the discriminative ability. An iterative algorithm is performed to derive the color space quantization and color feature generation. In order to illustrate the effectiveness of the proposed color representation, a specific image classification task, i.e., differentiating the adult images from benign ones, is employed. Experimental results show that our color feature achieves better classification performance and better efficiency compared with the traditional color histogram.
Bibliography:color histogram, mutual information, image classification
For many image classification tasks, color histogram is usually employed as an important ＂signature＂ to describe the color distribution of the image and infer the image content. However, most traditional color histograms cannot achieve satisfactory results in many image classification systems. In order to improve the accuracy and reduce the computational complexity of the classification task, an information-based color feature representation is proposed in this paper. The mutual information between the feature and the class label is adopted to evaluate the discriminative power of the feature. A novel quantization scheme is presented, which removes the redundant color components and combines the adjacent components together to generate a new feature to maximize the discriminative ability. An iterative algorithm is performed to derive the color space quantization and color feature generation. In order to illustrate the effectiveness of the proposed color representation, a specific image classification task, i.e., differentiating the adult images from benign ones, is employed. Experimental results show that our color feature achieves better classification performance and better efficiency compared with the traditional color histogram.
31-1943/U
WANG Shi-lin, LIEW Wee-chung Alan （1. School of Information Security Engineering, Shanghai Jiaotong University, Shanghai 200240, China 2. School of Information and Communication Technology, Gold Coast Campus, Criffith University, Queensland 4222, Australia）
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ObjectType-Article-2
ObjectType-Feature-1
ISSN:1007-1172
1995-8188
DOI:10.1007/s12204-011-1167-7