Abstract
This paper presents a computationally efficient haze removal and image enhancement methods. The major contribution of the proposed research is two-fold: (i) an accurate atmospheric light estimation using principal component analysis, and (ii) learning-based transmission estimation. To reduce the computational cost, we impose a constraint on the candidate pixels to estimate the haze components in the sub-image. In addition, the proposed method extracts modified haze-relevant features to estimate an accurate transmission using random forest. Experimental results show that the proposed method can provide high-quality results with a significantly reduced computational load compared with existing methods. In addition, we demonstrate that the proposed method can significantly enhance the contrast of low-light images according to the assumption on the visual similarity between the inverted low-light and haze images.
Highlights
As various digital cameras are introduced in the consumer market with related multimedia services, the enhancement of outdoor hazy images attracts increasing interest
The performances of haze removal methods were compared in terms of atmospheric light estimation, dehazing, and processing time
This paper presented haze removal and low-light image enhancement methods using the supervised learning manner
Summary
As various digital cameras are introduced in the consumer market with related multimedia services, the enhancement of outdoor hazy images attracts increasing interest. It is not easy to implement the method in consumer devices because of the high computational complexity in refining the transmission To solve this problem, Zhu et al proposed a linear model to estimate the depth using the color attenuation prior which represents the relationship between the brightness and saturation components of hazy images [2]. Fattal et al proposed a haze removal method based on the assumption that pixels having the same color in the haze region lie on the same line in the RGB color space [7] This method uses a local image formation model to estimate the accurate transmission, it provides an over-dehazed result according to the light condition. The contribution of the proposed method is two-fold: (i) estimation of atmospheric light vectors using the reduced pixels based on the PCA, and (ii) estimation of the transmission component using random forest via modified haze-relevant features.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
