The invention relates to a maximum entropy-based road shadow suppression method and belongs to the technical field of
image processing. The maximum entropy-based road shadow suppression method is used for reducing interference of road shadow during road detection. For solving the problem that similar methods are not thorough in shadow suppression and are prone to eliminating effective road areas, the technical scheme of the maximum entropy-based road shadow suppression method comprises, firstly, transferring a colorful image into an HSV (
hue, saturation and value)
color space, and extracting the HSV feature (log (H / S), log (V / S)) of every pixel point; secondly, projecting the HSV features of all the pixel points in the HSV image into a one-dimensional space, calculating the entropy values of the HSV features in the one-dimensional space, changing the
projection angle a to obtain the entropy values corresponding to different projection angles, and determining the
projection angle am corresponding to the maximum entropy value; lastly, reconstructing the HSV image on the basis of the
projection angle am, performing binarization and hole filling sequentially on the reconstructed image to obtain a shadow-suppressed road image. The maximum entropy-based road shadow suppression method is efficient, feasible and high in adaptability to different road conditions.