Large-space-range oil storage tank extraction method

An extraction method and oil storage tank technology, applied in the direction of neural learning methods, instruments, biological neural network models, etc., can solve the problems that the distribution consistency of training samples is greatly affected, and the background knowledge requirements of researchers are relatively high, so as to avoid radiation correction and outliers, fast automatic extraction, and strong universal applicability

Active Publication Date: 2021-07-09
MINISTRY OF ECOLOGY & ENVIRONMENT CENT FOR SATELLITE APPL ON ECOLOGY ENVIRONMENT
View PDF11 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this type of method still requires a lot of manual extraction of oil tank features, which requires relatively high background knowledge for researchers, and the robustness of the trained model is affected by the distribution of remote sensing image features and the distribution of training samples used in model training. Consistency has a big impact

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
  • Large-space-range oil storage tank extraction method
  • Large-space-range oil storage tank extraction method
  • Large-space-range oil storage tank extraction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0083] Obtain the tank extraction model;

[0084] Retrieve the multi-spectral high-resolution image of Gaofen-1 (GF1) in Dongying City, Shandong Province (March 14, 2020), such as figure 2 As shown in ; The multi-spectral high-resolution image of Gaofen 6 (GF6) in Dongying City, Shandong Province (September 22, 2019) was retrieved, such as image 3 As shown in ; The multi-spectral high-resolution image of No. 02 star (ZY3-02) (ZY3-02) (April 29, 2020) in Dongying City, Shandong Province was retrieved, as shown in Figure 4 shown in ; it is known that the spatial resolution of each image is 2 meters;

[0085] According to the distribution of oil storage tanks, different parts of different remote sensing images were intercepted. The image size of GF1 is 17557x14821 pixels; GF6 uses three images, the sizes are 5268x7301 pixels, 5032x7415 pixels and 5704x6845 pixels; Based on these images, the real distribution map of the oil tanks in the multi-spectral high-resolution image i...

Embodiment 2

[0108] Select the same multi-spectral high-resolution image as in Example 1 to obtain the oil storage tank extraction model;

[0109] In its obtaining process, the difference from Example 1 is:

[0110] The model used in the process of obtaining available samples is the Attention U-net neural network construction model. Correspondingly, the model for flushing training after obtaining available samples is also the Attention U-net neural network construction model;

[0111] The structural diagram of the Attention U-net neural network is as follows Figure 8 As shown in , the attention module is set in the neural network, and the feature map formed in the middle is further screened through the attention module;

[0112] After obtaining the oil storage tank extraction model, perform the same verification steps as in Example 1, and the satellite multispectral high resolution images retrieved in S1 are all completely consistent: the obtained average IOU, average producer accuracy a...

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 large-space-range oil storage tank extraction method, and the method comprises the steps: reading whether oil storage tanks are contained in a satellite image or not, monitoring the number, scale and spatial distribution of the oil storage tanks, and achieving the precise positioning of a VOCs emission source, thereby providing an effective technical support for the treatment of ozone and volatile organic compounds. The oil storage tank in the image is quickly identified by constructing an oil storage tank extraction model, and in the process of constructing the oil storage tank extraction model, the number of samples is increased through overturning transformation, random cutting and noise interference methods, so that the accuracy, practicability and universality of the model are improved.

Description

technical field [0001] The invention relates to a method for identifying and extracting an oil storage tank in a satellite image, in particular to a method for extracting an oil storage tank in a large space range. Background technique [0002] Oil storage tank extraction based on optical images is mostly based on the characteristics of image spectrum, texture geometry and shape. The methods are mainly divided into two categories: (1) Pixel-based extraction methods: take each pixel as the research object, count the spectral and texture features of each pixel in the image, and set The threshold is extracted from the storage tank. However, this type of method only considers the characteristics of a single pixel, and lacks the overall attributes of the objects to which the pixel belongs, as well as the spatial relationship between pixels. The extracted oil tanks are mostly trivial, and are greatly affected by the imaging angle and lighting conditions of the image. . (2) Obje...

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): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06V20/176G06V10/40G06V10/58G06F18/214G06F18/2415
Inventor 王玉于博杨晓钰张玉环马鹏飞周春艳张大为李巍孙军姜腾龙孙林张连华赵少华王中挺
Owner MINISTRY OF ECOLOGY & ENVIRONMENT CENT FOR SATELLITE APPL ON ECOLOGY ENVIRONMENT
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