A Remote Sensing Extraction Method of Ground Object Information Based on the Idea of ​​Vegetation Index

Abstract

The present invention relates to the technical field of remote sensing classification, in particular to a remote sensing method for extracting feature information based on the idea of ​​vegetation index, comprising the following steps: S1. Selecting target features and calculating corresponding NEW‑NDVI or NEW‑RVI values; S2 .Using threshold segmentation technology to extract the part greater than T in the value of NEW-NDVII or NEW-RVI obtained in step S1 respectively, can remove part of the non-target feature information; Calculate the value of NEW‑NDVI or NEW‑RVI to obtain the final target feature information. The ground object information remote sensing extraction method based on the idea of ​​vegetation index in the present invention extracts target information according to the characteristic information curve of the target surface object by analogy to the chlorophyll spectrum characteristic curve, the extraction accuracy is stable, the universality is strong, it is applicable to various environments, and the degree of automation is high, avoiding The interference of human factors in the extraction process is eliminated. At the same time, the extraction method is simple and the extracted information is accurate. It is of great significance to grasp the spatial distribution information of the target object and promote the application and development of remote sensing classification technology.

Description

technical field [0001] The invention relates to the technical field of remote sensing classification, in particular to a remote sensing extraction method of ground object information based on the idea of ​​vegetation index. Background technique [0002] Feature recognition and extraction have important application significance in many fields of society. Remote sensing is a large-scale, rapid acquisition of surface features information technology. Therefore, the technology and method of timely and accurately obtaining the spatial distribution information of ground objects is one of the important research directions in the field of remote sensing. At present, there are many remote sensing extraction techniques and methods for ground objects, such as extraction methods based on different classification techniques, including supervised classification and unsupervised classification (such as visual interpretation, decision tree, neural network, support vector machine, aggregatio...

AI Technical Summary

Problems solved by technology

[0003] However, according to the analysis of existing research results, it is found that the existing remote sensing extraction methods of ground objects have the following problems: first, the extraction accuracy is unstable, especially the spatial position accuracy is low; second, the method has regional limitations and poor universality; third, it needs Manually collect a large number of training samples of ground objects; fourth, the interference of human factors is relatively large; fifth, the method is complicated; sixth, the degree of automation is low

Images