We present what we believe to be a novel method based on pseudo-Wigner–Ville distribution (PWVD) and Rényi entropy for salient targets detection. In the foundation of studying the statistical property of Rényi entropy via PWVD, the residual entropy-based saliency map of an input image can be obtained. From the saliency map, target detection is completed by the simple and convenient threshold segmentation. Experimental results demonstrate the proposed method can detect targets effectively in complex ground scenes.
© 2010 Optical Society of America
Original Manuscript: September 9, 2009
Revised Manuscript: December 19, 2009
Manuscript Accepted: December 21, 2009
Published: February 4, 2010
Yuannan Xu, Yuan Zhao, Chenfei Jin, Zengfeng Qu, Liping Liu, and Xiudong Sun, "Salient target detection based on pseudo-Wigner-Ville distribution and Rényi entropy," Opt. Lett. 35, 475-477 (2010)