Fast approximation of trilateral filter for tone mapping using a signal processing approach

• Published in: Signal processing Vol. 89; no. 5; pp. 901 - 907
• Main Authors: Shen, Jianbing, Fang, Shiaofen, Zhao, Hanli, Jin, Xiaogang, Sun, Hanqiu
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.05.2009; Elsevier
• Subjects: Algorithms; Applied sciences; Approximation; Exact sciences and technology; Fast approximation; Image contrast; Information, signal and communications theory; Mapping; Mathematical analysis; Miscellaneous; Sampling, quantization; Signal and communications theory; Signal processing; Strategy; Telecommunications and information theory; Tone mapping; Trilateral filter; Fast approximation; Trilateral filter; Tone mapping; Signal processing; Mapping; Sampling; Algorithm; Fast filter; Kernel method
• ISSN: 0165-1684; 1872-7557
• DOI: 10.1016/j.sigpro.2008.11.009

In this paper, we present a novel fast approximation of the trilateral filter for high dynamic range (HDR) image tone mapping using a signal processing approach. Our fast trilateral filter-based tone-mapping algorithms are about 15–25 times faster than the existing one in Choudhury and Tumblin [The trilateral filter for high contrast images and meshes, in: Eurographics Symposium on Rendering ’03, 2003, pp. 186–196]. The novel tone-mapping algorithm for HDR images is presented based on the fast approximation of the gradient bilateral filter and the down-sampling strategy of computing the titling angle image. We show the comparison results with the fast approximation of the trilateral filter using full-kernel, truncated-kernel, piecewise-kernel, separable-kernel, respectively. The proposed approach is suitable for fast HDR image tone-mapping, and the experimental results show satisfactory performancce.