A Scene Classification Method for Remote Sensing Images

Abstract

The invention discloses a remote sensing image scene classification method. According to the distribution of local invariant feature points in the image, the method of the present invention pre-classifies the image into two types: uniform distribution of feature points and uneven distribution of feature points; then, for images with uniform distribution of feature points, the color histogram is used to The global features fused with texture features are trained and classified, and the images with uneven distribution of feature points are trained and classified using ScSPM (Sparse Coding Spatial Pyramid Matching, spatial pyramid matching model features based on sparse coding) local features. Compared with the prior art, the invention reduces the time required for classification while improving the classification accuracy.

Description

technical field [0001] The invention relates to a remote sensing image scene classification method. Background technique [0002] In recent years, high-resolution earth observation technology has developed rapidly, and has played an important role in many fields such as land survey, urban planning, disaster management and military affairs. Compared with low- and medium-resolution remote sensing images, high-resolution remote sensing images can provide detailed ground information, but the spatial structure distribution of various ground objects is more complex. With the improvement of the resolution of remote sensing images, the amount of spatial information is more abundant, and the geometric information and texture information of spatial objects are more obvious. Reduced heterogeneity and other issues. Therefore, although high-resolution remote sensing images provide a more detailed description of the surface, the difficulty of intelligent and automated information extrac...

AI Technical Summary

Problems solved by technology

[0004] Although the above-mentioned existing technologies have their own characteristics, they all have the contradiction between classification accuracy and classification time.

Images