Abstract
Under low illumination, the colour constancy of human vision can be used for correctly determining the colour of objects according to the fixed reflection coefficient of external light and objects. However, video image acquisition equipment does not implement the colour constancy characteristic of the human visual system. Under low illumination, only a small amount of light is reflected from the surface of the imaged object; as a result, the captured image is underexposed. After statistical analysis of low-light images, these inverted underexposed images appear foggy. Inversion is a uniform and reversible operation that is performed on the entire image. Hereby, a method is proposed for resolving low-light images using conventional physical models. First, a low-light image is inverted for obtaining a foggy image. Subsequently, a pyramid-type dense residual block network and a dark channel prior K-means classification method are applied to the foggy image, to calculate the transmission and atmospheric light. Finally, the parameters obtained from this solution are incorporated into the low-light imaging model to obtain a clear image. We subjectively and qualitatively analysed the experimental results, and used information entropy and average gradient for objective quantitative analysis. We demonstrate that the algorithm improves the overall brightness and contrast of the imaged scenes, and the obtained enhanced images are clear and natural.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: 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.