China Science Journal

DOI: 10.3724/SP.J.1004.2013.02064

Acta Automatica Sinica (自动化学报) 2013/39:12 PP.2064-2070

Haze Removal and Enhancement Using Transmittance-dark Channel Prior Based on Object Spectral Characteristic

SHI De-Fei¹ LI Bo¹ DING Wen¹ CHEN Qi-Mei¹
1.School of Electronic Science and Engineering, Nanjing University,Nanjing,210046,China

Abstract：
Images of outdoor scenes are usually degraded because of haze, fog and so on. Considering the atmospheric scattering model and object spectral characteristic, we expand dark channel prior and compute the transmittance channels in the LMS model to reduce corrections between RGB channels, and we also propose an adaptive LIPC algorithm to enhance the haze removal images based on logarithmic image processing (LIP). Results show that dark channel prior acts better in the transmittance channels than in RGB channels because of its lower correction with each other, and that haze removal images enhanced with LIPC are more credible and acceptable.

Key words：Image haze removal, object spectral characteristic, LMS, dark channel prior, LIPC

ReleaseDate：2014-07-21 17:04:34

Funds：National Natural Science Foundation of China (61105015), Natural Science Foundation of Jiangsu Province (BK2011511), and Science and Technology Support Program of Jiangsu Province (BE2011747)

PDF