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

SAR image change detection method based on deep learning and SIFT features

A technology of image change detection and deep learning, applied in the field of image processing, can solve problems such as low precision, unaccounted for, and sensitivity to speckle noise in SAR images, achieve strong adaptability, overcome low robustness, and improve precision

Active Publication Date: 2019-04-23
XIDIAN UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantages of this method are, firstly, this method is sensitive to the speckle noise of the SAR image, resulting in low precision of the final change detection
The disadvantage of this method is that the method does not consider the influence of SAR image speckle noise during joint classification, resulting in insufficient reliability of the selected training sample points

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
  • SAR image change detection method based on deep learning and SIFT features
  • SAR image change detection method based on deep learning and SIFT features
  • SAR image change detection method based on deep learning and SIFT features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The present invention will be further described below in conjunction with the accompanying drawings.

[0040] refer to figure 1 , the specific implementation steps of the present invention are as follows:

[0041] Step 1, read in the SAR image.

[0042] Read in two registered and rectified SAR images I and J in different phases of the same area.

[0043] Step 2, normalization.

[0044] According to the following formula, the SAR images I and J are normalized to obtain the normalized SAR image:

[0045]

[0046]

[0047]Among them, I' represents the normalized SAR image of SAR image I, min represents the minimum value operation, max represents the maximum value operation, and J' represents the SAR image after SAR image J normalization.

[0048] Step 3, construct training features.

[0049] The translation-invariant feature transformation SIFT method is used to extract the translation-invariant feature transformation SIFT feature S of two normalized SAR images ...

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 SAR image change detection method based on deep learning and SIFT features, which mainly solves the problem that the prior art is sensitive to the speckle noise of the SAR image, resulting in low precision of the final change detection result. Concrete steps of the present invention are as follows: (1) read in SAR image; (2) normalize; (3) structure training feature; (4) training depth neural network; (5) logarithm to two SAR images read in Ratio operation to obtain the log ratio difference image; (6) Construct the neighborhood feature sample matrix of the log ratio difference image; (7) Detect the log ratio difference image; (8) Output the change detection result map. The invention fully utilizes the stability characteristic of the SIFT feature on the speckle noise of the SAR image, overcomes the influence of the speckle noise of the SAR image, and improves the accuracy rate of SAR image change detection.

Description

technical field [0001] The present invention belongs to the technical field of image processing, and further relates to a synthetic aperture radar SAR (Synthetic Aperture Radar) based on deep learning and translation-invariant feature transformation SIFT (Scale-invariant Feature Transform, SIFT) features in the technical field of remote sensing image change detection. , SAR) image change detection method. The invention can be used to detect the changing area of ​​the SAR image in the same area at different time periods. Background technique [0002] Radar imaging technology was developed in the 1950s, and it has achieved leapfrog development so far. At present, this technology has been widely used in military affairs, agriculture, ocean, geology, surveying and mapping, etc. [0003] As an active microwave sensor, synthetic aperture radar has the characteristics of high resolution, all-weather, all-weather work and strong penetration. Therefore, SAR is not affected by atmos...

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): G06K9/46G06K9/62
CPCG06T7/0002G06T2207/20081G06T2207/10044G06V10/462G06F18/21375
Inventor 焦李成张丹汤志强马晶晶尚荣华马文萍赵进赵佳琦杨淑媛侯彪王爽
Owner XIDIAN UNIV
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