Supercharge Your Innovation With Domain-Expert AI Agents!

SAR Image Classification Method Based on Deep Feature Learning and Watershed

A classification method and deep feature technology, applied in the field of image processing, to achieve effective classification, improve contrast, and enhance robustness

Active Publication Date: 2018-06-15
XIDIAN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The present invention aims to solve the defects of the existing SAR image classification technology for mining middle-level features, and proposes a SAR image classification method based on deep learning and watershed

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 Method Based on Deep Feature Learning and Watershed
  • SAR Image Classification Method Based on Deep Feature Learning and Watershed
  • SAR Image Classification Method Based on Deep Feature Learning and Watershed

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] refer to figure 1 , the specific implementation steps of the present invention are as follows:

[0029] Step 1. Carry out watershed over-segmentation on the input SAR image to obtain the segmentation class label L={L 1 , L 2 ,...,L M}, where M is the total number of blocks divided by the watershed, L m is the class label of the mth sub-block, m∈[1,M].

[0030] Over-segmentation means that the value of M is much larger than the number of categories of the image, but it can guarantee that all pixels in each sub-block must belong to the same category of the image.

[0031] The specific process of this step is as follows:

[0032] (1a) Gamma (Gamma) filtering is performed on the input SAR image to obtain a smoothed image I;

[0033] (1b) Calculate the gradient map G of the classic edge detection operator (prewitt operator) of I 1 ;

[0034] (1c) for the gradient map G 1 Perform reconstruction to obtain the gradient map G 2 , the reconstruction rule is G 1 Set the t...

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 classification method based on deep feature learning and watershed, which belongs to the field of image processing technology, and mainly solves the problem that middle and low-level features are applied to SAR image classification, which is prone to scene misclassification, poor regional consistency, and burrs on the boundary. question. The classification process is as follows: calculate the watershed over-segmentation class label L for the input SAR image; calculate the gamma (Gabor) feature F1 of the input SAR image; input the Singular Value Decomposition (KSVD) algorithm to obtain the training dictionary D after sampling F1; F1 and D perform convolution and maximum pooling to obtain the convolution feature F2; then input F2 into the sparse autoencoder to obtain the deep feature F3; input F3 into the SVM for classification, and obtain the classification result R1; in the watershed segmentation As a result, each sub-block position of L performs voting statistics on R1 to obtain the final classification result. The invention has the advantages of fast operation speed, accurate edge classification and good regional consistency, and can be used for SAR target recognition.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to a method related to SAR image classification, which can be applied to target recognition. Background technique [0002] Synthetic Aperture Radar (SAR) is a high-resolution microwave imaging radar for all-weather and all-weather detection and reconnaissance of the earth. SAR can effectively identify camouflage and penetrate cover, so it has been widely used in military and civilian fields such as remote sensing mapping, military reconnaissance, and earthquake relief. SAR images have the characteristics of rich texture information and strong multiplicative coherent speckle noise. In the imaging process of SAR images, different ground features have different back reflection and scattering characteristics, thus presenting different textures, so SAR images often contain rich texture information. On the other hand, since SAR adopts a coherent imaging system, the imaged SAR image is in...

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): G06K9/62G06V10/764G06V20/13
CPCG06F18/23213G06F18/2411G06V20/13G06V10/449G06V10/54G06V10/764
Inventor 侯彪焦李成刘贺姚若玉马晶晶马文萍张涛刘闯
Owner XIDIAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com