Abstract
Image representation is an important branch of machine vision. Appropriate representation is the premise and foundation to obtain good performance of image classification. Color characteristic is one of the important and readily available characteristics of image. However, the color has always been a controversial characteristic in the research of image processing and classification. Hence, in this paper, we present an effective multiple representation fusion method for color image classification to verify color information can improve the accuracy of image classification. Our method first employs three original color component images of RGB to generate the corresponding three kinds of new pixels images. The new pixels images and original color component images of the subject are complementary in the representing the subject, so we design a non-parameter weight fusion function for integration them to improve the accuracy of image classification. The experimental results on the AR, GT, IMM and COIL100 datasets prove that our proposed method outperforms than the gray image classification. This paper undoubtedly proves the significance of color component in image classification.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
