Small sample remote sensing target detection method and system based on weight dictionary learning

Abstract

The invention discloses a small sample remote sensing target detection method and system based on weight dictionary learning. The method comprises the steps of acquiring remote sensing image data to be classified, substituting the data into a pre-trained target detection model to obtain a target category corresponding to a remote sensing image, and obtaining the target detection model through small sample data based on weight dictionary learning training. According to the method, a lightweight small-sample remote sensing target detection model is constructed by adopting a weight dictionary learning mode, so that the number of learnable parameters can be effectively reduced, overfitting of the model during training under small data is prevented, and the small-sample learning performance ofthe model is improved. The knowledge learned by the model in a source domain can be well reserved, so that the problem of disastrous forgetting is avoided. The remote sensing target detection method based on the weight dictionary provided by the invention has good generality, can be used for improving other remote sensing target detection models based on deep learning, and improves the small sample learning ability of the remote sensing target detection models.

Description

technical field [0001] The invention relates to remote sensing image target detection, in particular to a small-sample remote sensing target detection method and system based on weight dictionary learning. Background technique [0002] The automatic remote sensing image target detection technology can automatically locate and identify the interesting target in the static remote sensing image. The target detection method of remote sensing image based on deep learning has achieved rapid development, but this kind of target detection method of remote sensing image based on deep learning still has certain limitations. [0003] The target detection model of remote sensing images based on deep learning relies on a large number of training samples. These models can only achieve good performance after tens of thousands or more training iterations on a large number of training samples. When the training samples are insufficient, these models are prone to overfitting, and the perform...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem that the remote sensing target detection model based on deep learning relies on a large amount of training data and has poor scalability for new tasks, the present invention provides a small-sample remote sensing target detection method based on weight dictionary learning, including:

Images