Abstract
Most methods for the detection and removal of specular reflections suffer from nonuniform highlight regions and/or nonconverged artifacts induced by discontinuities in the surface colors, especially when dealing with highly textured, multicolored images. In this paper, a novel noniterative and predefined constraint-free method based on tensor voting is proposed to detect and remove the highlight components of a single color image. The distribution of diffuse and specular pixels in the original image is determined using tensors’ saliency analysis, instead of comparing color information among neighbor pixels. The achieved diffuse reflectance distribution is used to remove specularity components. The proposed method is evaluated quantitatively and qualitatively over a dataset of highly textured, multicolor images. The experimental results show that our result outperforms other state-of-the-art techniques.
© 2014 Optical Society of America
Full Article | PDF ArticleMore Like This
Hui-Liang Shen and Qing-Yuan Cai
Appl. Opt. 48(14) 2711-2719 (2009)
Hui-Liang Shen and Zhi-Huan Zheng
Appl. Opt. 52(19) 4483-4493 (2013)
Zhenmin Zhu, Chao Yang, Hailiang Zhang, Fumin Zhang, and Xuemei Liu
Appl. Opt. 59(23) 6873-6880 (2020)