Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method and system for extracting target object from image

A target object and image technology, applied in the field of remote sensing image processing, can solve the problems of multi-detailed features, discrete distribution of building targets, and different sizes, and achieve the effect of improving extraction accuracy and extraction effect

Pending Publication Date: 2021-06-18
CAPITAL NORMAL UNIVERSITY
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Compared with natural scene images, building targets in remote sensing images have the characteristics of discrete distribution, complexity, different sizes, and multi-detail features. Traditional semantic segmentation methods are not directly applicable to building semantic segmentation in remote sensing images.
The deep learning methods used in recent years, such as DANet and PAN, do not take into account the differences, salience and multi-level fusion correlation between different channels in the model.

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
  • Method and system for extracting target object from image
  • Method and system for extracting target object from image
  • Method and system for extracting target object from image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] figure 1 It is a flow chart of the target object extraction method of the present invention. Such as figure 1 As shown, the present invention provides a method for extracting a target object from an image, the method comprising the following steps:

[0043] S1: Receive an input image and a target object extraction request, where the extraction request is used to indicate the target object to be extracted.

[0044] In this step, the input images may be remote sensing images, or other images in the form of pictures, which is not limited in the present invention.

[0045] When inputting an image, a target object extraction request can be input, and the extraction request is used to indicate the target object to be extracted. Specifically, the image can be partitioned and numbered, and the number can be used to indicate the area to be extracted, or the default method can be used to Indicating the regions to be extracted is only an example, and other methods that can be i...

Embodiment 2

[0093] Figure 10 It is a schematic diagram of the target object extraction system of the present invention. Such as Figure 10 As shown, the present invention also provides a system for extracting a target object from an image, the system comprising:

[0094] a receiving unit, configured to receive an input image and a target object extraction request, and the extraction request is used to indicate the target object to be extracted;

[0095] The extraction unit is used to call the semantic segmentation deep learning model to extract the feature information of the target object in the image;

[0096] The output unit is configured to output the image of the target object based on the extracted feature information of the target object.

[0097] Preferably, the semantic segmentation deep learning model includes a feature extraction network module, the feature extraction network module has a channel grouping-based horizontal connection residual network structure, and the featur...

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 relates to a method and system for extracting a target object from an image, and the method comprises the following steps: S1, receiving an input image and a target object extraction request which is used for indicating a target object needing to be extracted; s2, calling a semantic segmentation deep learning model, and extracting feature information of a target object in the image; and S3, outputting an image of the target object based on the extracted feature information of the target object. According to the invention, accurate end-to-end building extraction and segmentation can be realized, and the extraction precision and the extraction effect of the target object are improved.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, in particular to a method and system for extracting a target object from an image. Background technique [0002] The rapid and efficient extraction of architectural objects using high-resolution remote sensing images is the basis for applications such as land resource management, fine-grained mapping, land use change monitoring, and human settlement suitability assessment. However, high-resolution images also bring problems such as large amount of calculation, complex calculation process, and partial information redundancy. In addition, there are problems such as multi-scale space, structural complexity, large differences in distribution, and complex surroundings of buildings. , causing certain difficulties and challenges to the efficient extraction of building information in high-resolution images. [0003] Currently, building extraction algorithms can be divided into me...

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): G06K9/00G06K9/34G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/176G06V10/267G06V10/462G06N3/045G06F18/253
Inventor 张振鑫李振钟若飞陈思耘
Owner CAPITAL NORMAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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