Sample expansion method and system based on foreground and background feature fusion

Abstract

The invention discloses a sample expansion method and system based on foreground and background feature fusion. The method comprises the steps: dividing a remote sensing ground feature classificationdata set into a source data set and a target data set based on ground feature categories; constructing a small sample source ground object classification task based on the source data set, and training a feature extractor, a hybrid model and a classifier based on the small sample source ground object classification task; constructing a small sample target ground object classification task based onthe target data set; performing sample expansion by using the trained feature extractor and the hybrid model based on the target classification task; wherein each task comprises a first task and a second task; the mixed feature is a feature synthesized by a foreground feature and a background feature by using a mixed model; according to the method, the hybrid model is trained based on the classification task, additional manual annotation is not added to expand the training sample, the training cost is reduced, the trained feature extractor and the hybrid model are utilized to expand the target data set, the classifier is trained, and the sample expansion method is realized.

Description

technical field [0001] The invention relates to the field of remote sensing intelligent recognition, in particular to a sample expansion method and system based on fusion of foreground and background features. Background technique [0002] Few-shot learning is to quickly learn a model from a small amount of data. Few-shot image recognition is a challenging emerging field in artificial intelligence. Traditional deep learning methods require a large amount of training data. When the amount of data is relatively small, it is easy to overfit the model and reduce performance. However, there is a shortage of data in many fields, such as the identification of rare species, medical diagnosis pictures, etc. Due to the difficulty in obtaining data or labeling, traditional deep learning techniques cannot be well applied in these fields. [0003] With the development of aerospace technology, my country's earth observation system has been initially formed. We can obtain high-resolutio...

AI Technical Summary

Problems solved by technology

[0008] In order to solve the problem of additional training cost in expanding samples through the data generation network in the prior art, the present invention provides a sample expansion method based on fusion of foreground and background features, including:

Images