DML-GANR: Deep Metric Learning With Generative Adversarial Network Regularization for High Spatial Resolution Remote Sensing Image Retrieval

TL;DR

This paper introduces DML-GANR, a novel deep metric learning method with GAN regularization, designed to improve high-resolution remote sensing image retrieval accuracy when training data is limited.

Contribution

It proposes a new deep metric learning framework with GAN regularization to prevent overfitting and enhance retrieval performance on small datasets.

Findings

DML-GANR outperforms existing methods on three datasets.

The approach effectively mitigates overfitting in small-sample scenarios.

High-level feature extraction combined with GAN improves retrieval accuracy.

Abstract

With a small number of labeled samples for training, it can save considerable manpower and material resources, especially when the amount of high spatial resolution remote sensing images (HSR-RSIs) increases considerably. However, many deep models face the problem of overfitting when using a small number of labeled samples. This might degrade HSRRSI retrieval accuracy. Aiming at obtaining more accurate HSR-RSI retrieval performance with small training samples, we develop a deep metric learning approach with generative adversarial network regularization (DML-GANR) for HSR-RSI retrieval. The DML-GANR starts from a high-level feature extraction (HFE) to extract high-level features, which includes convolutional layers and fully connected (FC) layers. Each of the FC layers is constructed by deep metric learning (DML) to maximize the interclass variations and minimize the intraclass variations. The generative adversarial network (GAN) is adopted to mitigate the overfitting problem and validate the qualities of extracted high-level features. DML-GANR is optimized through a customized approach, and the optimal parameters are obtained. The experimental results on the three data sets demonstrate the superior performance of DML-GANR over state-of-the-art techniques in HSR-RSI retrieval.