Leveraging Generative Deep Learning Models for Enhanced Change Detection in Heterogeneous Remote Sensing Data | IEEE Conference Publication | IEEE Xplore

Leveraging Generative Deep Learning Models for Enhanced Change Detection in Heterogeneous Remote Sensing Data


Abstract:

In this paper, we introduce an innovative approach for Change Detection (CD) in heterogeneous (multimodal) multi-temporal remote sensing (RS) images employing deep featur...Show More

Abstract:

In this paper, we introduce an innovative approach for Change Detection (CD) in heterogeneous (multimodal) multi-temporal remote sensing (RS) images employing deep features comparison through the utilization of two advanced deep learning models: Generative Adversarial Networks (GANs) and autoencoders. First, Deep Convolutional GANs are implemented to convert multimodal image(s) into synthetic image(s) of the same modality. Subsequently, autoencoders are trained and employed to extract compressed representations of both initial images. Finally, a change map is obtained by combining/fusing the original image with its corresponding generated change-free image resulting from the difference between the two learned compressed representations. Our proposed CD technique can accommodate CD algorithms for RS images expressing any type of change. Experimental evaluations on very high-resolution optical and Synthetic Aperture Radar (SAR) imagery validate the enhanced performance of the proposed method compared to existing state-of-the-art CD techniques in handling heterogeneous RS data.
Date of Conference: 08-11 July 2024
Date Added to IEEE Xplore: 11 October 2024
ISBN Information:
Conference Location: Venice, Italy

Funding Agency:


References

References is not available for this document.