Construction Method of Similarity Matrix in ncut Segmentation of Ultrasonic Image
A similarity matrix, ultrasonic image technology, applied in image analysis, image data processing, instruments, etc.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0054] figure 2 (a) is an ultrasound image with a size of 240x240 (unit: pixel), which is preprocessed by anisotropic diffusion and homomorphic filtering in sequence. The preprocessing results are shown in figure 2 (b) and figure 2 (c); A simple linear iterative clustering algorithm is used to divide the preprocessed image into a plurality of subregions with irregular boundaries, namely superpixels, wherein the number K of pixels contained in each small region is 700, See figure 2 (d); calculate the average gray value of each sub-region and the Euclidean distance of the average gray value between the sub-regions; calculate the texture feature vector of each sub-region and use the chi-square distance to represent the dissimilarity between the sub-regions, The two are combined according to formula (7) to obtain a new similarity matrix, and then verify the validity of the similarity matrix, and use the Ncut algorithm to cluster these sub-regions into two categories. The clu...
Embodiment 2
[0056] image 3 (a) is an ultrasound image with a size of 240x240 (unit: pixel), which is preprocessed by anisotropic diffusion and homomorphic filtering in sequence. For the results of anisotropic diffusion filtering, see image 3 (b), the results of homomorphic filtering are shown in image 3 (c), using a simple linear iterative clustering algorithm to divide the preprocessed image into multiple sub-regions with irregular boundaries, namely superpixels, wherein the number K of pixels contained in each small region is 700, See image 3 (d); Calculate the average gray value of each sub-region and the Euclidean distance of the average gray value between the sub-regions; calculate the texture feature vector of each sub-region and use the Jeffrey divergence distance to represent the difference between the sub-regions Similarity, the two are combined according to formula (7) to obtain a new similarity matrix, and then the validity of the similarity matrix is verified, and the ...
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