首页    期刊浏览 2024年12月03日 星期二
登录注册

文章基本信息

  • 标题:A NEW ILLUMINATION INVARIANT FEATURE BASED ON FREAK DESCRIPTOR IN RGB COLOR SPACE
  • 本地全文:下载
  • 作者:SIOK YEE TAN ; HASLINA ARSHAD ; AZIZI ABDULLAH
  • 期刊名称:Journal of Theoretical and Applied Information Technology
  • 印刷版ISSN:1992-8645
  • 电子版ISSN:1817-3195
  • 出版年度:2016
  • 卷号:93
  • 期号:1
  • 出版社:Journal of Theoretical and Applied
  • 摘要:Mobile markerless augmented reality (AR) requires a descriptor that can act in real time and at the same time robust to illumination invariance. Several color descriptors have been proposed to increase illumination invariance such as RGB-SIFT. According to the previous study, FREAK descriptor has been identified as the most suitable descriptor for mobile markerless AR due to FREAK reviews less computation time compared to other descriptors. However, color information presents important information has been largely ignored by FREAK. Hence, this research proposed a new illumination invariant feature based on FREAK descriptor in RGB color space for mobile markerless AR. Red, green and blue color channel will applied separately to FREAK descriptor. Hence, there are four types of descriptor been introduce; Red-FREAK, Green-FREAK, Blue-FREAK and RGB-FREAK. The proposed color FREAK descriptors are evaluated using images of different illumination color from Amsterdam Library of Object Images data set. The results show that the proposed color FREAK descriptors have enhanced the accuracy under illumination invariance for mobile markerless AR.
  • 关键词:Markerless Augmented Reality; Feature Descriptor; RGB; Illumination Invariance; FREAK
国家哲学社会科学文献中心版权所有