Research on road shadow removal algorithm based on adaptive brightness elevation model
QIU Guoyong
LI Li
LI Liangfu
MA Weifei
Abstract：For the problems of brightness elevation model shadow removal algorithm GSR,such as the manual setting of key parameters and the defect in the division of the equal brightness regions and the luminance compensation method,an road shadow emoval algorithm SGRSR based on the adaptive brightness elevation model was proposed.Firstly,the influence of pavement crack and pavement texture for the division of the shadow region was eliminated by using morphological dilate operation and Gaussian smoothing filter.Secondly,the maximum entropy threshold segmentation algorithm was used to calculate the partition threshold value of the shaded area and the non-shaded area of the image after Gaussian smoothing.By this way,the threshold was determined automatically.Finally,the pavement shadow was eliminated by using the improvement of the equal brightness region partition model and the luminance compensation method.Experimental results showed that the SGRSA algorithm could eliminate the shadow of the pavement image automatically and had the more natural transition in the brightness of the pavement image compared with the GSR algorithm.
Keywords：adaptive brightness elevation modelshadow removalmaximum entropy threshold segmentationluminance compensation
Publication Date：2018-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：9( 79-87 )