Neighbor Propagation Clustering Image Segmentation Method Based on Fuzzy Connectivity

A technology of fuzzy connectivity and image segmentation, applied in the field of image processing, can solve the problem of not being able to give full play to the role of spatial features and density features, and achieve the effect of reducing the amount of data, overcoming the high spatial and temporal complexity, and improving the segmentation accuracy.

Active Publication Date: 2018-08-17
航遨航空科技(西安)有限公司
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to address the shortcomings of the traditional similarity calculation method that cannot give full play to the role of spatial features and density features, and propose an AP clustering image segmentation method that uses fuzzy connectivity to calculate similarity to improve the segmentation effect

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Neighbor Propagation Clustering Image Segmentation Method Based on Fuzzy Connectivity
  • Neighbor Propagation Clustering Image Segmentation Method Based on Fuzzy Connectivity
  • Neighbor Propagation Clustering Image Segmentation Method Based on Fuzzy Connectivity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] refer to figure 1 , the concrete steps of the present invention are as follows:

[0044] (1) Use the Normalized Cut superpixel technology to perform superpixel segmentation on the image, and use superpixels as data points. There is a parameter N in the Normalized Cut superpixel algorithm, which is used to guide the number of superpixels and is set to 1000.

[0045] (2) For each superpixel, extract its spatial features and density features.

[0046] The superpixel spatial feature extraction method is as follows: calculate the average value of x coordinates of all pixels inside the superpixel as the x coordinate of the superpixel, and the average value of y coordinates of all pixels as the y coordinate of the superpixel, and take the x and y coordinates of the superpixel as value as its 2D spatial feature.

[0047] The superpixel density feature extraction method is: map the original image to the LUV color space, and for each superpixel, calculate the average L color v...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The present invention discloses an affinity propagation clustering image segmentation method based on fuzzy connectedness. The method is mainly used for solving the problems of over segmentation and low segmentation accuracy of other similar methods, which are generated due to a large target span. The affinity propagation clustered image segmentation method based on fuzzy connectedness comprises the implementation steps of: (1) performing super pixel segmentation on an image; (2) extracting spatial characteristics and density characteristics of super pixels; (3) according to the spatial characteristics of the super pixels, calculating a neighboring relationship; (4) according to the density characteristics of the super pixels and the neighboring relationship, calculating an affinity relationship; (5) according to the affinity relationship, calculating the fuzzy connectedness among all the super pixels; (6) according to the fuzzy connectedness and a spatial characteristic relationship, calculating a similarity degree among the super pixels; and (7) completing clustering on the super pixels by using affinity propagation clustering, and generating a segmentation result. The affinity propagation clustering image segmentation method based on fuzzy connectedness is a full-automatic segmentation method, has good target consistency of segmentation and high segmentation accuracy, solves the problem of difficulty in determining bias parameters of an existing AP clustering image segmentation method, and has a good segmentation ability for a nature color image.

Description

technical field [0001] The invention belongs to the field of image processing, and relates to an image segmentation method, in particular to a fuzzy connection-based nearest neighbor propagation clustering image segmentation method, which can be used for automatic segmentation of natural color images. Background technique [0002] In 1996, Udupa proposed a theoretical framework of fuzzy connectivity, and proposed a threshold fuzzy connectivity algorithm for medical image segmentation. Since then, relative fuzzy connectivity algorithm and iterative relative fuzzy connectivity algorithm have been proposed one after another. The fuzzy connectivity method uses the spatial position characteristics of pixels to calculate the proximity relationship, uses other features such as grayscale or color and the proximity relationship to calculate the affinity relationship, and finally uses the fuzzy connectivity to determine the target to which the pixel belongs. The combination of differ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/00G06F17/30
Inventor 葛洪伟杜艳新杨金龙苏树智袁运浩
Owner 航遨航空科技(西安)有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products