Unmanned aerial vehicle image plant canopy information extraction method based on ensemble learning

Abstract

The invention provides an unmanned aerial vehicle image plant canopy information extraction method based on ensemble learning, and the method comprises the following steps: S1, obtaining an unmanned aerial vehicle visible light remote sensing image, and carrying out data preprocessing on the unmanned aerial vehicle visible light remote sensing image; S2, performing image segmentation by adopting an object-oriented method to complete object feature extraction; and S3, adopting a Stacking ensemble learning model to realize extraction of plant canopy information. According to the method, an object-oriented and Stacking integrated machine learning method is combined, so that the parameter consumption of model construction in an information extraction process is greatly reduced, and manpower, material resources and time are greatly saved; and moreover, the traditional mode that a single classifier is used for carrying out pattern recognition on the high-resolution remote sensing image is broken through, the method can give full play to the advantages of each algorithm model, is more excellent in generalization, stability and applicability, and greatly improves the extraction precision and speed of the plant canopy information based on the unmanned plane image.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image extraction, in particular to a method for extracting plant canopy information from drone images based on integrated learning. Background technique [0002] The plant canopy is the main place for photosynthesis of trees, the most direct part to reflect tree information, and the easiest part to obtain information in remote sensing images. By extracting plant canopy information from remote sensing images, it is possible to estimate the height, biomass, canopy density and volume of trees, crops and other plants, and to monitor canopy changes caused by pests, droughts, fires, etc. Therefore, timely and accurate acquisition of plant canopy parameter information is of great significance for sustainable agricultural development and forest health monitoring. Traditional surveys have a large workload, long cycle, and low efficiency. Satellite remote sensing is limited by low spatial resolution...

AI Technical Summary

Problems solved by technology

[0003] At present, a large number of scholars are discussing and researching on the extraction of plant canopy information based on UAV images. The current object-oriented classification methods are mostly supervised classification methods, which require more human-computer interaction operations in practical applications and are less dependent on manual experience. Strong; Most of the training samples required for supervised classification are manually selected, which is relatively time-consuming and labor-intensive, and most of them are limited to using a single classifier for pattern recognition of high-resolution remote sensing images. It is often difficult to accurately identify complex ground objects, resulting in long time-consuming, The accuracy is low, and for different regions, the versatility of different models under different data is poor; therefore, there is an urgent need for an integrated learning classification model with stronger generalization ability, higher prediction accuracy, and wider versatility

Images