Research Article
Night Time Image Enhancement by Improved Nonlinear Model
@INPROCEEDINGS{10.1007/978-3-319-73447-7_34, author={Yao Zhang and Chenxu Wang and Xinsheng Wang and Jing Wang and Le Man}, title={Night Time Image Enhancement by Improved Nonlinear Model}, proceedings={Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II}, proceedings_a={MLICOM}, year={2018}, month={2}, keywords={Night time images Brightness Adaptive Nonlinearity Model Otsu Threshold}, doi={10.1007/978-3-319-73447-7_34} }- Yao Zhang
Chenxu Wang
Xinsheng Wang
Jing Wang
Le Man
Year: 2018
Night Time Image Enhancement by Improved Nonlinear Model
MLICOM
Springer
DOI: 10.1007/978-3-319-73447-7_34
Abstract
Low light or poor shooting angle and other issues often make the camera to take night time images and affect the naked-eye observation or computer identification, so it is important to enhance the lightness of night time image. Although the existing non-linear luminance enhancement method can improve the brightness of the low light area, the excessive promotion led to high light area distortion. Based on the existing image luminance processing algorithm, we proposed an adaptive night time image improving method in the basis of nonlinear brightness enhancement model is proposed to process the segmentations of image brightness by using the logarithmic function. The segmentation threshold is determined by the Otsu, and the adjustment factor of the backlight region in the transfer function is calculated from the area ratio of the backlight area. The conclusion comes from the simulation. The method involves improving the image quality and ensuring that the entire picture is natural without distortion. In the meanwhile, the processing speed is not much slower compared with the existing processing algorithms.
