A Method for Analyzing the Effect of Landscape Features on the Accuracy of Remote Sensing Classification Spots

Abstract

The invention provides a method for analyzing the influence of landscape features on the precision of remote sensing classification spots, including step 1, acquiring data, including performing data standardization processing on original images; step 2, performing image recognition on the data obtained in step 1, including classifying and Post-classification processing, wherein the classification process includes the determination of the classification pattern; step 3, the correlation calculation between the classification pattern and the true value data pattern is carried out on the data obtained in step 2, and the regression curve fitting is established Regression model, including the definition of landscape index expressing landscape characteristics; step 4, statistically test the regression model established in step 3, and evaluate the classification error expressed by the landscape index. The present invention adopts the theoretical basis of the accuracy evaluation and analysis based on the spatial features of the remote sensing classification, uses the landscape features of the remote sensing classification map to describe and express the classification accuracy, and provides the basis and guidance for the application and research related to the land cover thematic map.

Description

【Technical field】 [0001] The invention relates to the field of remote sensing image classification, in particular to a method for analyzing and evaluating the influence of landscape features on the accuracy of remote sensing classification map spots. 【Background technique】 [0002] Remote sensing classification thematic maps are widely used in many fields, such as land cover / use change monitoring, animal habitat site selection, hydrological analysis, risk analysis, and natural resource surveys. They are often used to describe the spatial distribution and shape of land cover, estimate land Coverage and other aspects (Stehman and Czaplewski, 1998; Smith et al., 2002). The accuracy and error distribution of thematic maps directly affect the application range and effect of thematic maps (Foody, 2002; Smith et al., 2002; Smith et al., 2003). Therefore, it is of great significance to fully understand the source and spatial distribution range of land cover map errors for land cove...

AI Technical Summary

Problems solved by technology

However, these studies all take a single pixel as the research object, only considering the possible impact of the pixel position, ignoring the shape characteristics and spatial distribution structure characteristics of the ground object itself.

According to general experience, in the classification results of remote sensing, the shape and scale distribution of the classification map may cause a large difference in the error distribution.

Images