Semantic image segmentation method based on conditional random field graph structure learning

A conditional random field, image segmentation technology, applied in the field of image understanding in computer vision, can solve the problem of poor segmentation effect and achieve good segmentation effect

Active Publication Date: 2018-07-20
ZHEJIANG UNIV OF TECH
View PDF5 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the shortcomings of poor segmentation effect of existing semantic image segmentation methods, the present invent

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
  • Semantic image segmentation method based on conditional random field graph structure learning
  • Semantic image segmentation method based on conditional random field graph structure learning
  • Semantic image segmentation method based on conditional random field graph structure learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The present invention will be further described below.

[0046] A method for semantic image segmentation based on conditional random field graph structure learning, said method comprising the following steps:

[0047] 1) Train a fully convolutional neural network or use an off-the-shelf fully convolutional neural network for rough segmentation of semantic images;

[0048] 2) Use the rcf neural network to learn the conditional random field graph structure, the process is as follows:

[0049] At present, some people have used richer convolutional features (RCF) to detect object boundaries in images. This embodiment uses the same deep neural network to learn CRF graphs by fine-tuning network parameters on CRF graph data. We first describe the structure of the network, which is actually a modification of the VGG16 network. Modifications include: 1) Cut all fully connected layers and pool5 layers; 2) Each conv layer in VGG16 is connected to a conv layer with core size 1x1...

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

Provided is a semantic image segmentation method based on conditional random field graph structure learning. The method includes the following steps: 1) training a fully convolutional neural network or adopting an existing fully convolutional neural network to perform semantic image coarse segmentation; 2) using an rcf neural network to learn a conditional random field graph structure; 3) trainingconditional random field model parameters through the graph structure obtained by learning and using a conditional random field model obtained through training to perform semantic image segmentation,wherein the step of semantic image segmentation is as follows: solving a Maximum A Posteriori inference problem and finding an optimal label of x by calling an alpha-beta extension routine. The invention provides the semantic image segmentation method with a good segmentation effect.

Description

technical field [0001] The invention belongs to the field of image understanding in computer vision, and relates to a method for semantic segmentation of semantic images. Background technique [0002] Semantic image segmentation (pixel labeling) is an important task in pattern recognition. This problem has been studied extensively in computer vision and many techniques have been developed, among which methods based on conditional random fields (CRF) are crucial because they 1) combine rich features (whether learned or 2) smooth segmentation boundaries and contrast-sensitive potentials; 3) the ability to model label consistency in local regions. With such strength, segmentation with CRF clearly outperforms results without CRF, especially when the local feature representation is weak. Contents of the invention [0003] In order to overcome the disadvantage of poor segmentation effect of existing semantic image segmentation methods, the present invention provides a semantic...

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): G06T7/143G06T7/10G06N3/08G06N3/04
CPCG06N3/08G06T7/10G06T7/143G06N3/047G06N3/045
Inventor 王振华丁福光郭东岩张剑华刘盛陈胜勇
Owner ZHEJIANG UNIV 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