Hyperspectral image classification method based on three-dimensional convolutional neural network

A hyperspectral image and three-dimensional convolution technology, applied in the field of hyperspectral image classification, can solve the problems of destroying the spatial information and spectral information of three-dimensional hyperspectral images, rearranging rough three-dimensional signals, and being unable to fully utilize the spatial information of hyperspectral images, etc. Achieving a good classification effect

Inactive Publication Date: 2018-02-02
HARBIN INST OF TECH
View PDF5 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the rough rearrangement of three-dimensional signals into two-dimensional signals by the existing two-dimensional convolutional neural network, which not only cannot make full use of the spatial information in t

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
  • Hyperspectral image classification method based on three-dimensional convolutional neural network
  • Hyperspectral image classification method based on three-dimensional convolutional neural network
  • Hyperspectral image classification method based on three-dimensional convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0029] Specific implementation mode 1: Combination figure 1 To explain this embodiment, the specific process of the hyperspectral image classification method based on the three-dimensional convolutional neural network in this embodiment is as follows:

[0030] Step 1: Import the hyperspectral image data set into the MATLAB platform, and normalize the three-dimensional data information in the hyperspectral image data set imported into the MATLAB platform layer by layer to obtain a processed data set;

[0031] Select the same amount of data (200 for soil, 200 for water and 200 for sky) of all categories (soil, water and sky) tags from the processed data set and record the space coordinates;

[0032] Use the processed data set as the test set;

[0033] The three-dimensional data information in the hyperspectral image data set includes spectral information and spatial information;

[0034] The hyperspectral image data set is in the form of a three-dimensional matrix;

[0035] Step 2: Take th...

Example Embodiment

[0044] Specific embodiment two: this embodiment is different from specific embodiment one in that in the first step, the hyperspectral image data set is imported into the MATLAB platform, and the three-dimensional data information in the hyperspectral image data set imported into the MATLAB platform is sorted layer by layer. One process to obtain a processed data set; the specific process is:

[0045] Import the hyperspectral image data set into the MATLAB platform used in the experiment, and perform layer-by-layer normalization on the three-dimensional data information of the hyperspectral image data set imported into the MATLAB platform. The formula is:

[0046]

[0047] In the formula, 1≤i≤W, 1≤j≤L, 1≤k≤H, Is the normalization function, Is the three-dimensional data of the hyperspectral image data set at positions i, j, and k, i, j represent the spatial position of the three-dimensional data information in the hyperspectral image data set, k represents the spectral band of the ...

Example Embodiment

[0051] Specific embodiment three: This embodiment is different from specific embodiments one or two in that in the third step, a three-dimensional convolutional neural network is built on the Keras platform according to a training set in the form of a three-dimensional matrix; the specific process is:

[0052] 1) The three-dimensional convolution layer performs sliding window convolution on the training set in the form of the three-dimensional matrix as the input of the three-dimensional convolutional neural network, and uses the three-dimensional convolutional layer as the first layer of the three-dimensional convolutional neural network;

[0053] When using a three-dimensional convolutional layer as the first layer of a three-dimensional convolutional neural network, input_shape (when the training set in the form of a three-dimensional matrix is ​​used as input) parameters must be provided;

[0054] 2) The channel position of the training set data after sliding window convolution is...

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 hyperspectral image classification method based on a three-dimensional convolutional neural network and relates to a hyperspectral image classification method. The inventionaims to solve a problem that an existing two-dimensional convolutional neural network coarsely rearranges a three-dimensional signal into a two-dimensional signal, spatial information in a hyperspectral image which can not be fully utilized, and the spatial information and spectral information in an original three-dimensional hyperspectral image are destroyed. The method comprises the steps of (1)inputting a hyperspectral image data set into a MATLAB platform and obtaining a processed data set, (2) taking a new hyperspectral image as a training set, (3) building the three-dimensional convolutional neural network according to the training set of a three-dimensional matrix form, (4) using the training set of the three-dimensional matrix form to train the three-dimensional convolutional neural network and obtaining a trained three-dimensional convolutional neural network, and (5) using a test set of a three-dimensional matrix form to input the trained three-dimensional convolutional neural network and obtaining a test set classification result. The method is used in the field of image classification.

Description

technical field [0001] The invention relates to a hyperspectral image classification method. Background technique [0002] The development direction of hyperspectral image classification technology is very rich, and the convolutional neural network has been found to be very suitable for the classification of hyperspectral images in recent years. However, the traditional convolutional neural network used for hyperspectral classification mostly uses two-dimensional Convolutional neural networks are used to classify hyperspectral images with distinct three-dimensional properties. This method will roughly rearrange the 3D signal into a 2D signal, which not only fails to make full use of the spatial information in the hyperspectral image, but also destroys the spatial and spectral information in the original 3D hyperspectral image. Contents of the invention [0003] The purpose of the present invention is to solve the rough rearrangement of three-dimensional signals into two-d...

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
IPC IPC(8): G06K9/62G06N3/04
CPCG06N3/045G06F18/241G06F18/214
Inventor 林连雷周祝旭杨京礼
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap