The present invention provides an improved
image enhancement method. The method comprises: S1, performing normalization
processing of image pixel
gray level f(x, y) to obtain n(x, y), wherein an improved normalization
processing mode is employed: (
img file='dest_path_image001. TIF' wi='255' he='183' / ); S2, performing coding of parameters to be optimized (
img file='dest_path_image002. TIF' wi='37' he='71' / ), randomly generating one group of initial individuals to form an initial
population, and inputting a control parameter
crossover probability pc, a
mutation probability pm, a
population size N, the maximum operation algebra G and the like; S3, determining whether an
evolution algebra t is equal to the G or not, if the
evolution algebra t is equal to the G, finishing the
algorithm, and outputting the optimal solution of the (
img file='dest_path_image003. TIF' wi='37' he='88' / ), or else, turning to the next step; S4, employing a roulette strategy to select M individuals, and performing
crossover and
mutation operation of the individuals according to the
crossover and
mutation method in the genetic manipulation; S5, selecting two vaccines, individual number to be inoculated and the
inoculation site number to perform immunization operation, making out immunization selection after
inoculation, and employing the optimal individual reservation strategy of the
population after the
inoculation; and S6, allowing one group (img file='dest_path_image004. TIF' wi='37' he='71' / ) to correspond to one non-linear transformation function F(u), and employing a non-linear transformation function to perform image gray
scale transformation to obtain an output image g (x, y).