Shadow removing method for color image and application

A shadow removing method for an image, comprising a shadow-free feature analysis and extraction process, a shadow-free transformation parameter acquisition process and a shadow-free feature imaging process;

• the method specifically comprising the following steps:
• (1) the shadow-free feature analysis and extraction process:
• carrying out statistical analysis on pixel RGB values of the same material in the original image under different illumination conditions, or finding out invariable characteristics in a shadowy image through a physical imaging model to obtain the expression K=f(R, G, B, . . . ) of shadow-free features, wherein K is a shadow-free feature matrix, R, G and B correspond to three component matrices of the red color, green color and blue color of the original image respectively, . . . represents other parameters, and f is a conversion expression, wherein the linear relation between the B and G components is selected to represent the shadow-free features of the image for road surface detection, and the expression of the shadow-free features is:
• f
• ⁢
• :
• ⁢
• ⁢
• (
• G
• ,
• B
• ,
• b
• )
• ->
• K
• ,
• K
• =
• G
• -
• b
• B
• ,
• wherein b is other parameter to be determined, and the geometric meaning is the intercept of a fitting straight line, and for a road surface region, K is a fixed value, and K is the required shadow-free feature matrix;
• (2) the shadow-free transformation parameter acquisition process:
• for the shadow-free feature matrix K=f(R, G, B, . . . ) in step (1), obtaining the corresponding shadow-free feature matrix K through calculation by determining the parameters in the expression;
• and
• (3) the shadow-free feature imaging process:
• converting the shadow-free matrix into a grayscale image, and obtaining a shadow-free feature grayscale image through feature visualization.