Level Set Image Segmentation Method Based on Local Guided Kernel Fitting Energy Model
A technology of local guidance kernel and energy model, applied in the field of image processing, can solve the problems of time-consuming, poor image effect, complex grayscale unevenness, etc., to improve segmentation accuracy, avoid re-initialization problems, improve accuracy and segmentation efficiency Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0039] Concrete implementation steps of the present invention include as follows:
[0040] (1) Input the segmented image and set the initialization parameters: given scale parameter α, time step Δt, normalization parameter ε of Heavide function, symbolic distance function constant ρ, and covariance ξ;
[0041] (2) Initialize the level set function φ of the evolution curve, which is defined as the signed distance function φ(x,t)=0.
[0042] (3) calculate according to the curve evolution equation described in description step 4;
[0043] (4) Calculate the evolved level set function φ according to the Gaussian filter equation in step 5, that is, φ n =G ξ *φ n ;
[0044] (5) Judging whether the described level set evolution curve is satisfied and terminated, if yes, then output the images and segmentation results of each segmented region. Otherwise, the Gaussian filtered level set function φ n+1 = φ n As the initial level set function for the next iteration, go to step three....
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


