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

SAR image classification based on 2D-PCA and convolution neural network

A convolutional neural network and 2D-PCA technology, applied in the field of image processing and SAR image classification, can solve the problems of reduced classification accuracy, high noise sensitivity, poor model robustness, etc., to reduce dimensions, improve classification accuracy, The effect of reducing complexity

Inactive Publication Date: 2019-01-18
XIDIAN UNIV
View PDF2 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The shortcomings of this method are high training complexity, poor robustness of the model, and low recognition rate
The disadvantage of this method is that it is relatively sensitive to noise. When some noise is randomly added, the classification accuracy will be reduced.

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 classification based on 2D-PCA and convolution neural network
  • SAR image classification based on 2D-PCA and convolution neural network
  • SAR image classification based on 2D-PCA and convolution neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The realization scheme of the present invention is: select the training image and the test image and carry out amplification, use two-dimensional principal component analysis to reduce the dimensionality of the data after amplification, build the convolutional neural network model, use the training sample after dimensionality reduction to train convolution The neural network model finally classifies the test samples.

[0027] refer to figure 1 , the specific implementation steps of the present invention are as follows:

[0028] Step 1. Select training and test samples and perform amplification.

[0029] 1.1) From the MSTAR data set containing BMP2 armored vehicles, BTR70 armored vehicles, and T72 main battle tank data, a total of 1617 images with a depression angle of 17 degrees are selected as training images X 1 , each training image is randomly translated by (x, y) pixels in the horizontal and vertical directions, where x, y are integers within (-10, 10), repeating...

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 synthetic aperture radar image classification method based on two-dimensional principal component analysis and convolution neural network, which mainly solves the problems oflow classification accuracy, poor robustness, high complexity and high noise sensitivity in the prior art. The realization scheme is as follows: 1. selecting training and test samples, and amplify them; 2. carrying out two-dimensional principal component analysis is carried out for each sample in the training sample set and the test sample set to obtain the training sample set and the test sampleset after dimension reduction; 3. constructing the convolution neural network model and training it with the training sample after dimension reduction; 4. loading the trained model weights into the constructed convolution neural network, and inputting the reduced dimension test samples into the convolution neural network to complete the classification of the original SAR images. The invention reduces model training complexity and sensitivity to noise, improves model robustness and classification accuracy, and can be used for target detection.

Description

technical field [0001] The invention belongs to the field of image processing, and further relates to the technical field of SAR image classification, which can be used for target detection. Background technique [0002] Synthetic aperture radar is an all-weather, all-weather working radar system, which is widely used in military, agricultural, navigation, geographical surveillance and many other fields. With the development of radar technology, SAR image classification has become an important field. SAR is especially suitable for target classification, detection and surveillance. Due to the difference between the imaging mechanism of SAR images and traditional images, there are also great differences between SAR images and traditional images in the process of understanding and interpretation. Classification by human eyes is very time-consuming and labor-intensive, so it is necessary to study some algorithms to achieve the classification effect. With the development of de...

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): G06K9/62G06N3/04
CPCG06N3/045G06F18/2411G06F18/214
Inventor 侯彪李井亮焦李成马晶晶马文萍白静
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