Abstract | ||
---|---|---|
A new method is presented to estimate a dynamic atmospheric-light map in daytime haze-images, which output a brighter result and suppress noise-amplificationUsing computer-generated training samples instand of manual collections. Simples generated by this method has a smaller inter-class distance, which can significantly improve the accuracy of the regression model.Appropriate quality-related features are analyzed and selected, and the proposed block-based feature extraction significantly reduces the input dimensions. The running time and the restoration are improved simultaneously. Given that single image dehazing is an ill-posed problem, it can be challenging to control the enhancement of haze images. In this paper, we propose a fast and accurate dehazing algorithm based on a learning framework. Using randomly generated training samples, we tackle the difficult problem of sampling haze/clear image pairs. Seven haze-relevant features based on image quality are extracted and analyzed. A regression model is learned using support vector regression (SVR), which can estimate the transmission map accurately. Further, a new method is presented to estimate the dynamic atmospheric light, which improves the performance in the sky and shadow regions. Experimental results demonstrate that the proposed approach has a lower computational complexity, and the dehazing results are visually appealing even on extremely challenging photos, such as street views, thick fog, and sky regions. Subjective analysis and objective quality assessments demonstrate that, the proposed method generates superior results than the state-of-the-art methods. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1016/j.neucom.2017.03.024 | Neurocomputing |
Keywords | Field | DocType |
Image restoration,Dehazing,Dynamic atmospheric light,Learning-based,Regression | Computer vision,Shadow,Pattern recognition,Regression analysis,Support vector machine,Image quality,Feature extraction,Artificial intelligence,Sampling (statistics),Image restoration,Mathematics,Computational complexity theory | Journal |
Volume | Issue | ISSN |
245 | C | 0925-2312 |
Citations | PageRank | References |
8 | 0.46 | 29 |
Authors | ||
6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zhong Luan | 1 | 9 | 0.83 |
Yuanyuan Shang | 2 | 210 | 16.83 |
Xiuzhuang Zhou | 3 | 380 | 20.26 |
Zhuhong Shao | 4 | 90 | 9.65 |
Guodong Guo | 5 | 2548 | 144.00 |
Xiaoming Liu | 6 | 1627 | 93.31 |