Unlock instant, AI-driven research and patent intelligence for your innovation.

A mining area FVC calculation method and system with enhanced edge sampling and improved unet model

A mining area and model technology, applied in computing, computer components, neural learning methods, etc., can solve the problems of sparse vegetation distribution and difficult to identify, strong surface heterogeneity, etc., to achieve fast calculation speed, high inversion accuracy, and improved data. The effect of the fault

Inactive Publication Date: 2022-04-12
CHINA UNIV OF MINING & TECH (BEIJING) +3
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Aiming at the deficiencies in the prior art, the purpose of the present invention is to provide a mining area FVC calculation method that enhances edge sampling and improves the Unet model, uses UAV images to construct a high-resolution mining area vegetation image sample data set, and overcomes mining area scenarios. The subsurface has strong heterogeneity, and the distribution of vegetation is sparse and difficult to identify; introduce high-precision digital elevation model data, supplement the terrain feature information in the mining area scene, improve the multi-dimensional characteristics of the sample data, and provide methods and theories for the construction of mine vegetation sample data sets Base

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
  • A mining area FVC calculation method and system with enhanced edge sampling and improved unet model
  • A mining area FVC calculation method and system with enhanced edge sampling and improved unet model
  • A mining area FVC calculation method and system with enhanced edge sampling and improved unet model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0069] Such as Figure 1 to Figure 10 As shown, a mining area FVC calculation method that enhances edge sampling and improves the Unet model, the method is as follows:

[0070] A. Data collection of ground vegetation parameters in the mining area scene: determine the mining area research area, use drones to take overlapping orthophoto aerial photography of the mining area research area according to the mapping accuracy and obtain the orthophoto set. The minimum overlapping ratio of the aerial photography is 32-38% ( To design the flight plan, first survey the entire research area on the spot, clarify the terrain of the research area, conduct simulation planning for the take-off, landing, and flight of the UAV, and then divide the entire research area into multiple flights with overlapping areas on the device The survey area, the overlapping ratio of the area depends on the demand for data accuracy, preferably 35%), then image stitching is performed to obtain the orthophoto ima...

Embodiment 2

[0087] Such as Figure 1 to Figure 10 As shown, a mining area FVC calculation method that enhances edge sampling and improves the Unet model, the method is as follows:

[0088] A. Data collection of ground vegetation parameters in the mining area scene: Determine the mining area research area, use drones to take overlapping orthophoto aerial photography of the mining area research area according to the mapping accuracy and obtain an orthophoto set. The mapping accuracy is centimeter level, and the minimum overlapping ratio of aerial photography 32-38%, then image stitching is performed to obtain the orthophoto image of the research area, and then the spatial three-dimensional solution model SFM is established to generate dense point cloud data, which is transformed to obtain high-precision digital geographic elevation model data;

[0089] According to a preferred embodiment of this embodiment, in step A, according to the mapping accuracy of the orthophoto set, flight planning ...

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 mining area FVC calculation method that enhances edge sampling and improves the Unet model, A. ground vegetation parameter data collection in the mining area scene; B. constructing a sample data set based on the cross-overlapping method of expanding selected areas; C. constructing and training improvement Unet neural network model; D, using the improved Unet neural network model to calculate the vegetation coverage of the mining area. The present invention first collects remote sensing data and ground soil vegetation coverage by unmanned aerial vehicle flight, and lays out ground control points to build a vegetation-related data acquisition system, which provides a data basis for vegetation coverage calculations, and then utilizes the alternate overlapping sampling method of expanding selected areas to perform training sample data Carry out segmentation and extraction to build a vegetation coverage sample database, and finally use the improved Unet network model to carry out model training and build a vegetation coverage network relationship model, and then accurately deduce the centimeter-level information data of vegetation coverage, which provides information for mining ecological environment monitoring management and mining development. The plan provides strong data support.

Description

technical field [0001] The invention relates to the fields of mining, artificial intelligence, ecology, remote sensing and geographic information, and in particular to a mining area FVC calculation method with enhanced edge sampling and improved Unet model. Background technique [0002] The mining of the mining area seriously affects the growth of vegetation. Timely and accurate detection of the vegetation growth in the mining area, and the use of vegetation coverage data to study the vegetation in the mining area can clarify the distribution and growth of the vegetation in the mining area ecosystem, and provide data for the subsequent management and mining of the mining area. At present, the methods of obtaining vegetation coverage by remote sensing monitoring can be mainly divided into the following three categories: 1. Statistical regression model: by establishing the linear relationship between the measured vegetation coverage and remote sensing information (Graetz 1988, ...

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): G06T17/05G06F30/10G06F30/27G06K9/62G06N3/04G06N3/08G06Q50/02G06V10/774G06V10/764G06V10/82
CPCG06T17/05G06F30/10G06F30/27G06N3/08G06Q50/02G06N3/045G06F18/214G06F18/241
Inventor 张成业邢江河郭添玉李全生郭俊廷张凯泽仁卓格佘长超李军
Owner CHINA UNIV OF MINING & TECH (BEIJING)