Super-pixel segmentation method based on SLIC algorithm

A superpixel segmentation and superpixel technology, applied in image analysis, calculation, image enhancement and other directions, can solve the problem that the SLIC method is not suitable for the superpixel segmentation method, and achieve the effect of improving the processing speed.

Inactive Publication Date: 2017-02-08
HARBIN INST OF TECH
View PDF2 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the existing SLIC method is not suitable for the superpixel segmentation method, and propose a superpixel segmentation method based on the SLIC algorithm

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
  • Super-pixel segmentation method based on SLIC algorithm
  • Super-pixel segmentation method based on SLIC algorithm
  • Super-pixel segmentation method based on SLIC algorithm

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0019] The superpixel segmentation method based on the SLIC algorithm of the present embodiment, the SLIC algorithm refers to a simple linear iterative clustering image segmentation algorithm, combined with figure 1 Shown flow chart, described method is realized through the following steps:

[0020] Step 1, converting the multi-band image to the image of the CIE-LAB color space;

[0021] Step 2, according to the number of superpixels set, initialize the clustering center in the image of CIE-LAB color space;

[0022] Step 3: Disrupt the cluster centers within the set n*n field range to form new cluster centers;

[0023] Step 4. According to the calculation of the set distance between each point and the new cluster center, assign matching points to each new cluster center;

[0024] Step 5. Calculate the distance L1 between the new cluster center and the previous cluster center, and judge whether the value of L1 is less than the set threshold,

[0025] If so, return to step 3 ...

specific Embodiment approach 2

[0028] Different from the first embodiment, the superpixel segmentation method based on the SLIC algorithm of the present embodiment, the process of converting the multi-band image to the image of the CIE-LAB color space described in the first step is,

[0029] Step 11, invert the value of K by formula (1):

[0030] Y 0 = K ∫ λ S ( λ ) y ¯ ( λ ) d λ - - - ( 1 )

[0031] At this time, according to the Y in the XYZ space 0 A constant requirement of 100 is required, so let Y 0 =100; Obtain S(λ) value and Then deduce the value of K; in the formula, K represents the parameter used for normalization; Y 0 Represents the Y component of the CIE1931XYZ color space, and the Y component represents the green primary color; λ represents the w...

specific Embodiment approach 3

[0044] The difference from Embodiment 1 is that in the superpixel segmentation method based on SLIC algorithm in this embodiment, the spectral reflectance or spectral transmittance R(λ) of the object described in Steps 1 and 2 is obtained through normalization operation.

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 invention discloses a super-pixel segmentation method based on SLIC algorithm, and belongs to the field of image processing. A conventional SLIC method is not suitable for the super-pixel segmentation. The method comprises the steps: enabling a multiband image to be converted into an image in a CIE-LAB color space; initializing clustering centers in the image in the CIE-LAB color space according to the set number of super-pixels; disorganizing the clustering centers in a set n*n range, and forming new clustering centers; carrying out the calculation according to the set distances between each point and the new clustering centers, and distributing a point for each new clustering center; calculating the distance values L1 between the new clustering centers and the former clustering centers, judging whether the distance values L1 are greater than a set threshold value or not: returning to step and disorganizing the clustering centers again if the distance values L1 are greater than the set threshold value, or else, carrying out the iterative calculation of the distance values L1 between the new clustering center and the former clustering centers till the distance values L1 are less than the set threshold value, and obtaining a segmentation result.

Description

technical field [0001] The invention relates to a superpixel segmentation method based on SLIC algorithm. Background technique [0002] In recent years, the spatial resolution and spectral resolution of imaging hyperspectral have been getting higher and higher, which has played a great role in various fields such as agricultural automation and urban planning, and has great development potential. At present, the resolution of hyperspectral images has reached the sub-meter level, and the original methods are also difficult to apply in the fields of image recognition. [0003] Superpixel segmentation is widely used in the preprocessing of high-resolution images to obtain better processing results. The superpixel method saves the target outline, processes multiple element sets as an object, reduces the amount of calculation, obtains faster calculation speed, and better processing results. [0004] The currently used SLIC superpixel segmentation method is generally proposed for...

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 Applications(China)
IPC IPC(8): G06T7/11G06K9/62
CPCG06T2207/10036G06T2207/10032G06F18/232
Inventor 谷延锋王腾飞
Owner HARBIN INST OF TECH
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap