Transfer learning classification method for high spatial resolution remote sensing images based on openstreetmap

Abstract

The transfer learning classification method for high spatial resolution remote sensing images based on OpenStreetMap includes: automatically generating image object sample sets in the target domain based on OpenStreetMap data; using the source domain images with the same imaging sensor as the target domain images, and automatically generating them based on their historical classification maps The image object sample set in the source domain; the above target domain and source domain sample sets are combined to form a mixed sample set, which is used to train the migration learning algorithm classifier based on random forest; the final classifier is used to predict the image object type in the target domain, so as to obtain the final classification results. The present invention can extract the label of the target domain image object from the OpenStreetMap data without manually labeling the target domain image category. The classification of images reduces the classification cost and can be applied to the classification of large-scale high-spatial-resolution remote sensing images.

Description

technical field [0001] The invention belongs to the field of remote sensing image processing, in particular, relates to a high spatial resolution remote sensing image transfer learning classification method based on OpenStreetMap (OSM), the method can obtain a target domain sample set based on OpenStreetMap, combined with a source domain sample set, A transfer learning algorithm is used to classify high spatial resolution remote sensing images. Background technique [0002] Land surface information obtained from high spatial resolution remote sensing images can be used in urban planning, land monitoring and other industries. At present, obtaining surface type information from high-spatial-resolution remote sensing images is mainly divided into two categories: unsupervised classification and supervised classification. Supervised classification needs to prepare a sample set for classifier training in advance, while unsupervised classification does not require prior knowledge ...

AI Technical Summary

Problems solved by technology

For example, this type of method first needs to solve the problem of spatial position registration between map data and images. Spatial position offset often leads to the generation of wrong samples; the second problem is that open source map data like OpenStreetMap relies on public contributions. The completeness of data varies greatly in different regions. For example, the data completeness is higher in the coastal cities in the east of my country than in the west.

These problems bring challenges to directly rely on OpenStreetMap data to generate sample sets for classification

Images