A Parallel Extraction Method of Trees and Plants from UAV Images

An extraction method and unmanned aerial vehicle technology, applied in the field of investigation and research of forest plants, can solve the problems of low accuracy, long algorithm time-consuming, difficult to identify trees at the same time, and achieve high efficiency, speed and high precision

Inactive Publication Date: 2018-03-16
GUANGZHOU INST OF GEOGRAPHY GUANGDONG ACAD OF SCI
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, the accuracy of current digital image processing technology is generally low, and it is difficult to identify trees of different canopies at the same time
At the same time, due to the generally high resolution of current drone images, digital image processing requires a lot of calculations, resulting in a long time-consuming algorithm.

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 Parallel Extraction Method of Trees and Plants from UAV Images
  • A Parallel Extraction Method of Trees and Plants from UAV Images
  • A Parallel Extraction Method of Trees and Plants from UAV Images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] The present invention will be further described below in conjunction with specific embodiments.

[0015] The parallel extraction method of forest trees and plants for UAV images of the present invention, such as figure 1 Shown, including steps:

[0016] Step s101: Obtain a forest area image taken by a drone;

[0017] Step s102, using GPU parallel processing to perform binary large-scale object detection on the forest image based on scale space technology to obtain detection points;

[0018] Step s103: Screen the initial detection points, and delete the detection points that do not contain green;

[0019] Step s104: Use the screened detection points as the extraction result.

[0020] Using the PyCUDA platform, the parallel extraction algorithm of forest trees and plants from UAV images is realized. CUDA (ComputeUnified Device Architecture) technology is a C-like language GPU programming platform developed by NVIDIA. PyCUDA is a Python language package of CUDA, which provides auto...

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 method for parallel extraction of trees and plants from UAV images. Firstly, the method of GPU parallel processing is used to detect binary large-scale objects based on scale space technology on forest images captured by UAVs to obtain detection points, and then Use the CPU serial method to delete the detection points that do not contain green, and the remaining detection points are the extraction results. The scale-space technology adopted can identify trees of different canopy sizes at various scales with high precision. Moreover, the method of object detection is performed by using GPU parallel processing, which makes this extraction method have high efficiency and speed. Plant investigation and research provided support.

Description

Technical field [0001] The invention relates to the technical field of forest tree plant investigation and research, in particular to a method for parallel extraction of forest tree plants from drone images. Background technique [0002] UAV passive optical remote sensing has greater advantages than satellite passive optical remote sensing in terms of temporal and spatial resolution, availability, and cost. At present, more and more applications are used in forest tree plant investigation and research. For the extraction of forest trees (mainly orchards, plantations, etc.), digital image processing technology is mainly used to identify individual target plants from images taken by drones and perform statistics or analysis. [0003] However, the accuracy of current digital image processing technology is generally low, and it is difficult to identify trees in different canopies at the same time. At the same time, due to the generally high resolution of current UAV images, digital im...

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): G06K9/00
CPCG06V20/188
Inventor 姜浩李丹陈水森刘尉王重洋
Owner GUANGZHOU INST OF GEOGRAPHY GUANGDONG ACAD OF SCI
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