Coastline change analysis method based on deep learning

A technology of change analysis and deep learning, applied in the field of coastline change analysis, can solve problems such as different shoreline extraction criteria, complicated shoreline types, and difficulty in achieving accurate classification of shoreline types, and achieve enhanced effects and improved extraction accuracy.

Active Publication Date: 2021-11-09
HARBIN INST OF TECH
View PDF9 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the existing coastline types are complicated and it is difficult to realize accurate classification of coastline types; for different types of coastlines, the coastline extraction criteria are different; and given multi-temporal remote sensing images, how to realize accurate coastline classification in large Extracting the problem, a deep learning-based coastline change analysis method is proposed

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
  • Coastline change analysis method based on deep learning
  • Coastline change analysis method based on deep learning
  • Coastline change analysis method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0066] Specific implementation mode one: combine figure 1 Describe this embodiment, the specific process of a coastline change analysis method based on deep learning in this embodiment is:

[0067] Step 1. Select the coastline data set (multi-temporal coastline remote sensing image and corresponding coastline type (such as artificial coastline type, etc.)) of the coastline data set in the research area, and divide the coastline data set into a training set and a test set;

[0068] Construct a neural network, input the training set into the neural network for training, and obtain the trained network, input the test set into the neural network for testing, and output the coastline type. If the test results meet the requirements, a trained neural network is obtained. If the test results do not meet the requirements Then retrain by adjusting model parameters or training data sets until a trained neural network is obtained;

[0069] The training set includes multi-temporal coastli...

specific Embodiment approach 2

[0075] Specific embodiment two: the difference between this embodiment and specific embodiment one is: the coastline data set of the research area is selected in the step one, and the coastline data set is divided into a training set and a test set;

[0076] Construct a neural network, input the training set into the neural network for training, and obtain the trained network, input the test set into the neural network for testing, and output the coastline type. If the test results meet the requirements, a trained neural network is obtained. If the test results do not meet the requirements Then retrain by adjusting model parameters or training data sets until a trained neural network is obtained;

[0077] The specific process is:

[0078] Step 11, select the coastline data set of the research area, and divide the coastline data set into a training set and a test set, wherein the ratio of the training set to the test set is 4:1;

[0079] Step 1 and 2, construct VGG16, MobileNe...

specific Embodiment approach 3

[0083] Embodiment 3: This embodiment differs from Embodiment 1 or Embodiment 2 in that: in the step 2, the multi-temporal coastline remote sensing image is processed to obtain the processed multi-temporal coastline remote sensing image;

[0084] Input the processed multi-temporal coastline remote sensing image into the trained neural network to obtain the coastline type corresponding to the processed multi-temporal coastline remote sensing image; the specific process is as follows:

[0085] Perform image deaverage and clipping on the multi-temporal coastline remote sensing image to obtain the remote sensing image area of ​​interest;

[0086] Input the obtained remote sensing image area of ​​interest into a suitable trained neural network according to actual needs (for example, the MobileNet network can provide the fastest training speed, the Inception-ResNet network can provide the highest test set accuracy, etc.), and the processed The coastline types of multi-temporal coastl...

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 coastline change analysis method based on deep learning, and relates to a coastline change analysis method. The invention aims to solve the problems that the existing shoreline types are complicated, accurate classification of the shoreline types is difficult to realize, shoreline extraction criteria for different types of shorelines are different, and how to realize accurate extraction of the shoreline in a large scene for given multi-temporal remote sensing images. The method comprises the following steps: 1, selecting a coastline data set of a researched area; constructing a neural network to obtain a trained neural network; 2, obtaining a processed multi-temporal coastline remote sensing image, inputting the processed multi-temporal coastline remote sensing image into the trained neural network, and obtaining a coastline type corresponding to the processed multi-temporal coastline remote sensing image; step 3, based on the obtained coastline type, extracting a coastline to obtain a coastline image; and 4, detecting the position change of the coastline based on the obtained coastline image, and calculating the erosion or deposition rate of the coastline. The invention is applied to the field of coastline change analysis.

Description

technical field [0001] The invention relates to a coastline change analysis method. Background technique [0002] The coastline is the boundary between the sea and the land. Generally, there are natural and artificial coastlines. Due to the particularity of its geographical location and the impact of human development, the location of the coastline is always changing. The classification of coastlines is an important basis for its protection and development and management of marine resources. my country has a wide territorial sea, and the total length of coastlines is about 32,000 kilometers. Coastline tidal flats have good economic benefits, and also have development value in terms of ecology, environment, and geology, and are precious resources of the country. At the same time, the extraction and classification of coastlines play an important guiding role in the development of urban planning, economic development, land use and other fields. The classification results of c...

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): G06T7/12
CPCG06T7/12G06T7/136G06T5/30G06N3/08G06T2207/10032G06T2207/20081G06T2207/20084G06T2207/20132G06N3/045G06F18/24G06F18/214Y02A10/40
Inventor 刘天竹张献豪谷延锋
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