Image level set segmentation method based on local gray clustering characteristics
A level set segmentation and local grayscale technology, applied in image analysis, image enhancement, image data processing, etc., can solve the problem of inconsistency of image grayscale
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0063] An image level set segmentation method based on local gray-level clustering features, such as figure 1 shown, including the following steps:
[0064] Step 1: Read the image I to be segmented.
[0065] Step 2: Use the linear weighted sum of M orthogonal basis functions to fit the offset field b, and initialize the weight values of each basis function.
[0066] In this embodiment, M is 15, and the formula of the offset field b(x) is shown in formula (1):
[0067] b=w T G (1)
[0068] Among them, w=(w 1 ,w 2 ,...,w M ) T is the basis function weight column vector, T is the transpose operation symbol, G=(g 1 , g 2 ,..., g M ) T is a column vector composed of M basis functions, g 1 , g 2 ,..., g M is a pairwise orthogonal 4th-order Legendre polynomial function, w 1 =w 2 =...=w M =1 is the weight value of each basis function initialized.
[0069] Step 3: Initialize the level set function set of the image: according to the number N of regions to be divided ...
Embodiment 2
[0096] In the simplest case of the present invention, the image is divided into two regions N=2, and k=1 is calculated according to formula (2).
[0097] Such as Figure 8 In the situation shown, (a) is the four images to be segmented and the initial zero level set segmentation curve, (b) is the segmentation result of the four images to be segmented, where the level set segmentation curve smoothing coefficient v of the four images to be segmented takes values in sequence 0.1×255×255, 0.5×255×255, 0.3×255×255, 0.02×255×255.
Embodiment 3
[0099] The method of the present invention divides the real image with noise, offset field and weak boundary information, divides the image into two regions N=2, and calculates k=1 according to the formula (2). Such as Figure 9 As shown, (a) is the four images to be segmented and the initial zero level set segmentation curve, (b) is the offset field estimation of the four images to be segmented obtained by the method of the present invention, (c) is the image after four offset field corrections , (d) is the segmentation result of the four images to be segmented, where the coefficient v of the level set segmentation curve smoothing term of the four images to be segmented takes values 0.0045×255×255, 0.003×255×255, 0.005×255×255, 0.01 ×255×255.
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com