High resolution sar image change detection method based on incremental cae

An image change detection and high-resolution technology, applied in the field of image processing, can solve the problems of not fully utilizing the two-dimensional spatial information of SAR images, the noise suppression effect of high-resolution SAR images is not obvious, and the detection effect of SAR images is not good. Training time, the effect of suppressing noise in high-resolution SAR images, and enhancing robustness

Active Publication Date: 2019-07-23
XIDIAN UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that because the stacked autoencoder network can only process one-dimensional data, it does not make full use of the two-dimensional spatial information of the SAR image. SAR image detection effect is not good
However, the shortcomings of this method are that when the salient area is large, the pixel-based clustering method is still time-consuming, the noise suppression effect on high-resolution SAR images is not obvious, and the detection accuracy is not high

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
  • High resolution sar image change detection method based on incremental cae
  • High resolution sar image change detection method based on incremental cae
  • High resolution sar image change detection method based on incremental cae

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] The present invention will be further described below in conjunction with the accompanying drawings.

[0045] refer to figure 1 , the concrete steps that the present invention realizes are as follows:

[0046] Step 1, input high-resolution synthetic aperture radar SAR image.

[0047] Input two registered high-resolution synthetic aperture radar SAR images I of different time phases in the same area 1 and I 2 .

[0048] Step 2, construct the initial sample set.

[0049] For two high-resolution SAR images I 1 and I 2 Sliding window cutting is performed to obtain 17956 image blocks of 15×15 pixels as the initial sample sets D1 and D2 respectively.

[0050] Step 3, obtain the normalized sample set.

[0051] Through the linear scaling method, the values ​​of the initial sample sets D1 and D2 are normalized to [0, 1] respectively, and the normalized sample sets V1 and V2 are obtained.

[0052] The specific steps of the linear scaling method are as follows:

[0053] ...

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 high-resolution SAR image change detection method based on incremental CAE, which mainly solves the problems in the existing methods that the SAR image change detection is easily affected by speckle noise, and the detection accuracy is not high and the detection speed is slow. Concrete steps of the present invention are as follows: (1) input high-resolution SAR image; (2) construct initial sample set; (3) normalize; (4) construct incremental convolution self-encoding CAE network; (5) construct error function ; (6) Train the incremental convolutional self-encoding CAE network; (7) Obtain the feature difference map; (8) Obtain the change detection result map at the 15×15 pixel scale; (9) Obtain the change detection results at three different scales Figures; (10) Fusion of change detection result graphs of different scales; (10) Output of high-resolution synthetic aperture radar SAR image change detection result graphs. The invention has the advantages of good robustness to the noise influence of SAR image change detection, high detection precision and fast detection speed.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a high-resolution Synthetic Aperture Radar (SAR) image change detection method based on incremental convolution self-encoding CAE (Convolution Auto Encoder) in the technical field of remote sensing image change detection . The invention can realize the detection of different areas of two high-resolution SAR images obtained in different time phases, and can be widely used in the fields of ground object coverage and utilization, forest and vegetation coverage detection, natural disaster monitoring and evaluation, urban planning and the like. Background technique [0002] High-resolution synthetic aperture radar (SAR) image change detection refers to the process of qualitatively or quantitatively analyzing and determining the characteristics of land surface changes using multi-temporal remote sensing images of the same geographical area acquired at different times. B...

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 & AuthorityPatents(China)
IPC IPC(8): G06T7/00G06T7/136G06K9/32G06K9/46G06N3/04
CPCG06T7/0002G06T7/136G06T2207/10044G06T2207/20221G06V20/13G06V10/507G06V10/25G06N3/045
Inventor焦李成屈嵘孟繁荣张丹杨淑媛侯彪马文萍刘芳尚荣华唐旭古晶陈璞花
OwnerXIDIAN UNIV