Remote sensing image cloud detection method and device based on full convolutional neural network

A convolutional neural network, remote sensing image technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as low detection accuracy, and achieve the effect of deepening network depth, enriching detailed information, and improving accuracy

Active Publication Date: 2020-06-12
SHENZHEN INST OF ADVANCED TECH
View PDF8 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a remote sensing image cloud detection method and device based on a fully convolutional neural network, to at least solve the technical problem of low detection accuracy of the existing cloud detection method

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
  • Remote sensing image cloud detection method and device based on full convolutional neural network
  • Remote sensing image cloud detection method and device based on full convolutional neural network
  • Remote sensing image cloud detection method and device based on full convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] According to an embodiment of the present invention, a kind of remote sensing image cloud detection method based on fully convolutional neural network is provided, see figure 1 , including the following steps:

[0041] S101: select the RGB band of Fengyun Meteorological Satellite Remote Sensing Image to build a data set, and obtain a training set in the data set;

[0042] S102: build the SP-HRNet network model, the network model includes continuous parallel multi-resolution sub-network, repeated multi-scale fusion module and separable convolution combination module with depth;

[0043] S103: training set input network model is trained, obtains the parameter of network model, forms network parameter model;

[0044] S104: Using the network parameter model to perform cloud detection on the remote sensing image.

[0045] The remote sensing image cloud detection method based on the full convolutional neural network in the embodiment of the present invention selects the RGB...

Embodiment 2

[0071] According to another embodiment of the present invention, a kind of remote sensing image cloud detection device based on fully convolutional neural network is provided, see Figure 5 ,include:

[0072] Data set acquiring unit 201 is used to select the RGB band of Fengyun Meteorological Satellite Remote Sensing Image to build a data set, and obtains a training set in the data set;

[0073] The network model construction unit 202 is used to construct the SP-HRNet network model, and the network model includes continuous and parallel multi-resolution sub-networks, repeated multi-scale fusion modules and separable convolution combined modules with depth;

[0074] Training unit 203, is used for training set input network model and trains, obtains the parameter of network model, forms network parameter model;

[0075] The detection unit 204 is configured to use the network parameter model to detect remote sensing image clouds.

[0076] The remote sensing image cloud detectio...

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 relates to the field of remote sensing detection, in particular to a remote sensing image cloud detection method and device based on a full convolutional neural network. The method comprises the steps of selecting an RGB waveband of a wind cloud meteorological satellite remote sensing image to construct a data set, and obtaining a training set in the data set; constructing an SP-HRNet network model, wherein the network model comprises a continuous and parallel multi-resolution sub-network, a repeated multi-scale fusion module and a depth separable convolution combination module;inputting the training set into a network model for training to obtain parameters of the network model, and forming a network parameter model; and performing remote sensing image cloud detection by using the network parameter model. According to the method and the device, the sub-networks with multiple resolutions can be kept all the time, so that information is not lost in the feature extractionprocess of the image, the network depth is deepened, the depth separable convolution is combined, the feature extraction capability of the network is improved, the detail information of a detection result is enriched, and the cloud detection precision is improved.

Description

technical field [0001] The invention relates to the field of remote sensing detection, in particular to a remote sensing image cloud detection method and device based on a fully convolutional neural network. Background technique [0002] According to the global cloud cover data provided by the International Satellite Cloud Climate Program (ISCCP), clouds cover more than 60% of the Earth's surface. Therefore, remote sensing images are extremely susceptible to being blocked by clouds during the imaging process, resulting in spectral distortion of the original objects and objects, which has a great impact on the information extraction of images. The existing cloud detection methods can be roughly summarized as spectral threshold method, spatial texture analysis method, pattern recognition detection method and machine learning method. The cloud detection method based on spectral threshold is the method with the longest research history, which mainly extracts various spectral fe...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/13G06N3/045G06F18/214
Inventor 林创陈劲松李洪忠
Owner SHENZHEN INST OF ADVANCED 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