The Research And Application Of Image Segmentation Method Based On Level Set Theory

Due to The level set method can transform the evolution of plane curve into the evolutionof the surface in a higher dimensional space, which can effectively solve the problem about thetopology change of curve, therefore it become the important research direction in the field ofimage segmentation.According to the different of image information, the level set model mainly divided intotwo categories, the model based on the edge and the model based on region. In essence, thesetwo kinds of models respectively use the gradient information and regional information of theimage.The first kind of model includes the level set model based on distance regularized-DRLSE model,The second kind of model includes the level set model based on regionalcharacteristics, and the level set model based on the fitting of the extended area, respectivelyachieve the image segmentation algorithm based on these models, and through the specificexperiment results, we compare and analyze the advantages and disadvantages of each model.This paper proposes a variational level set model based on local clustering. The modelmainly use the local information of the image. The model first uses the K-mean clusteringalgorithm for intensity clustering in the local region of each image point, and defines a localclustering criterion function for the image intensities, then consider the whole image, define aglobal clustering criteria and make it achieve the minimum to make the whole imagesegmentation to achieve the best. Then the global clustering criterion as an energy items of thelevel set energy function, plus the regularization item, curve length and the penalty function,become the level set energy function. According to the formula, we can get the partialdifferential equation and discrete partial differential equation, and finally achieve the imagesegmentation algorithm based on this level set model. The specific experimental results provethat this model can solve the problem about the segmentation of images with intensity inhomogeneities.This paper use the medical images for the experimental object. We calculate the area errorrate, the minimum iterations and convergence time by comparing the image segmentationbased on CV model, RSF model and LCVLS model. The area error rate proves thesegmentation of LCVLS model is better than the CV model and RSF model, and the minimumiterations and convergence time prove that the segmentation efficiency of the model is higherthan the CV model and RSF model, the minimum iterations and convergence time. On thewhole, it demonstrates the LCVLS model for medical image segmentation is more effective byspecific experimental results.