Method for improving accuracy of classification and extraction of target ground objects in remote sensing image

Abstract

The invention provides a method for improving accuracy of classification and extraction of target ground objects in a remote sensing image. The method weighs the extraction accuracy and the extraction ratio of the target ground objects, and a decision-making trade-off interval is obtained through setting thresholds of the extraction accuracy and the extraction ratio of the target ground objects, so that the accuracy of classification and extraction of the target ground objects in the existing remote sensing image is greatly improved.

Description

technical field [0001] The invention relates to the technical field of classification and extraction of remote sensing images, in particular to a method for improving the accuracy of classification and extraction of target features in multi-spectral remote sensing images. Background technique [0002] The extraction of target features in remote sensing images by classification technology has a very wide range of applications in various fields such as agriculture, forestry, water conservancy, and disasters in our country. The thematic maps of the distribution of various crops, the thematic maps of forest land and tree species distribution, and the disaster mapping produced by classification technology provide powerful objective data support for government decision-making. [0003] At present, classification technologies mainly include manual visual interpretation and automatic software interpretation. Manual visual interpretation requires relatively high experience and profes...

AI Technical Summary

Problems solved by technology

But for this method, the error rate is not as low as possible, because the target ground object Non-target features The lower the error rate is, the lower the extraction rat

Images