Image set partitioning method based on sampling

A technology of sampling images and image segmentation, applied in image analysis, image data processing, instruments, etc., can solve problems such as laborious

Inactive Publication Date: 2012-08-29
NANJING UNIV
View PDF2 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the previous interactive image segmentation algorithms all directly segment a single image.
So if the image set contains a lot of images, it is very laborious to directly apply the previous single image segmentation method to segment an image set.
The joint segmen

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
  • Image set partitioning method based on sampling
  • Image set partitioning method based on sampling
  • Image set partitioning method based on sampling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] In order to display the intermediate results clearly and intuitively, the present invention aligns the corresponding intermediate and final results of each target image in all the graphs in columns.

[0063] figure 2 An example of segmenting an image set is given. There are 5 images in this image set, including 1 sample image (1st image in the first row) and 4 target images (4 images in the second row). First, the user uses the Lazysnapping algorithm to manually segment the sampled image to separate the foreground and background, and the result is the second image in the first row. The image set only needs to be segmented to obtain a satisfactory result. The selection of the sampling image and the manual segmentation by the user correspond to step 1 of the image segmentation part in the specific implementation steps. According to the result of the segmentation, the present invention can automatically segment the other 4 sheets without any other interaction from the u...

Embodiment 2

[0065] image 3 The image set has 9 images in total, 2 of which are sampled images ( image 3 The 1st and 3rd images in the first row), and the remaining 7 are the target images ( image 3 in the second row). The sampled image is manually segmented by the user, and the segmentation result is image 3 The 2nd and 4th pictures in the first row, calculated according to the similarity, image 3 The sampling image of the first picture in the first row is similar to the first to third target images in the second row, image 3 The sampled image of the 3rd image in the first row is similar to the target image of the 4th to 7th image in the 2nd row. For this image set, the image segmentation process of the present invention is the same as figure 2 The middle image set is similar. Similar sampling images can be calculated for each target image, and the initial probability of each target image can be calculated according to the similarity (results such as image 3 The 3rd line of...

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 an image set partitioning method based on sampling. The method comprises the following steps of: image selection: extracting characteristic vectors of color histograms of all images in an image set and performing characteristic vector clustering; extracting color distributions of the foreground and background of a sampled image; performing similarity matching on the sampled image with a target image; calculating the foreground probability of the target image based on the single sampled image; calculating the foreground probability of the target image; and constructing a Gibbs energy minimizing formula, resolving a result of a foreground or background, which corresponds to each pixel, through image partitioning, and partitioning the image according to the constructed Gibbs energy minimizing formula by using an image partitioning method. The method has the remarkable advantage that a large amount of user interaction required by foreground and background partitioning operation on each image in a large-scale image set can be reduced greatly.

Description

technical field [0001] The invention relates to a method for segmenting the foreground of all images in a data set containing arbitrary images, in particular to an algorithm for segmenting the foreground of all images based on simple manual interaction of certain images in the image set. Background technique [0002] Currently, interactive image segmentation algorithms greatly simplify the task of image foreground segmentation. However, the previous interactive image segmentation algorithms all directly segment a single image. Therefore, if the image set contains a lot of images, it is very laborious to directly apply the previous single image segmentation method to segment an image set. The joint segmentation algorithm (co-segmentation) can segment all the foregrounds of some image sets at the same time, but this type of algorithm has a strong assumption about the image set: the foreground of the image set should be the same, or at least have very similar Color distributi...

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
IPC IPC(8): G06T7/00
Inventor 郭延文付彦伟
Owner NANJING UNIV
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