Single image dehazing with bright object handling

• Published in: IET computer vision Vol. 10; no. 8; pp. 817 - 827
• Main Authors: Riaz, Irfan, Fan, Xue, Shin, Hyunchul
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 01.12.2016; Wiley
• Subjects: Benchmarking; block-level haze thickness estimation; bright surface intensity value; bright-object handling capability; Channels; dark channel prior; DCP; DCP failure categorisation; Failure; haloes reduction; Handling; Haze; image fusion; image resolution; image restoration; local average haziness value; Luminous intensity; nonobvious failure; obvious failure; pixel-level haze thickness estimation; reliability map; reliability-guided fusion; Research Article; single image dehazing; Strategy; transmission estimation; transmission value; bright-object handling capability; transmission value; reliability-guided fusion; image fusion; DCP failure categorisation; dark channel prior; haloes reduction; block-level haze thickness estimation; single image dehazing; nonobvious failure; local average haziness value; DCP; pixel-level haze thickness estimation; image restoration; obvious failure; transmission estimation; image resolution; reliability map; bright surface intensity value

This study addresses the shortcomings of the dark channel prior (DCP). The authors propose a new and efficient method for transmission estimation with bright-object handling capability. Based on the intensity value of a bright surface, they categorise DCP failures into two types: (i) obvious failure...