The invention relates to an
image segmentation method combining a cloud model and a
level set, which comprises the steps of S1, acquiring a
histogram of an image, and performing
curve fitting on the
histogram; S2, dividing valley value intervals through the fitted curve; S3, acquiring three digital characteristics Ex, En and He by using a reverse
Gaussian cloud
algorithm, and obtaining a foreground cloud model or a background cloud model through
software or cloud; and S4, constructing an energy function, and solving the energy function to obtain a segmentation result. The
image segmentation method combines a cloud model and a
level set algorithm, and de-
linearization processing is performed on boundaries in the image by using the cloud model, so that the
occurrence probability and the degree of a problem of instable convergence caused by manual intervention are reduced, and the convergence corresponding to a
level set function is enabled to be accelerated; and meanwhile, the operationof initializing the
level set function by the aid of a cloud model
algorithm effectively reduces the
noise sensitivity of the function and solves a problem of continuously initializing the
level set function in the segmentation process.