Conditional GAN-based Remote Sensing Target Image Generation Method

Haoyang Liu, Zhiyi Hu, Jun Yu, Shouyi GaoInternational Journal of Advanced Network Monitoring and Controls 2020 · DOI

Abstract

Traditional Generative Adversarial Networks (GANs) often struggle with uncontrollable generation processes and a lack of diversity in synthesized remote sensing images. This paper proposes a novel generative model based on background conditions and visual attention mechanisms. By employing a U-Net generator and a color-coded conditional mask, the model focuses on learning specific target features while ignoring non-essential information. Experimental results indicate that the proposed method achieves a Peak Signal-to-Noise Ratio (PSNR) of 18.512 and a Structural Similarity Index (SSIM) of 88.47%, significantly outperforming standard autoencoder-based models.

1. Introduction

Training robust deep learning models for remote sensing target detection requires vast amounts of diverse data. However, acquiring high-resolution satellite imagery is often expensive and logistically challenging. While traditional data augmentation (e.g., rotation, cropping) increases sample size, it does not introduce new structural diversity. Generative models like GANs offer a path forward, but standard architectures lack the precision required to place specific targets in precise locations. This study addresses these limitations by introducing a Conditional GAN (CGAN) framework tailored for remote sensing applications.

2. Proposed Methodology

The proposed method transitions from "black box" generation to a controllable process using spatial and categorical constraints.

2.1 Conditional Mask and Attention Mechanism

To provide precise control, the researchers implemented a color-coded conditional mask. Different RGB values in the mask correspond to specific target categories, such as standard ships or aircraft carriers. This mask serves as a visual attention mechanism, directing the network to synthesize targets at specific coordinates while preserving the surrounding environment.

2.2 Network Architecture

The generator utilizes a U-Net structure with eight layers of down-sampling and eight layers of up-sampling. Skip connections are integrated to pass low-level background information directly across the network, ensuring that the synthesized targets blend realistically with the sea texture without distorting the original background.

The Conditional GAN workflow: The generator (G) synthesizes a fake image G(x) from a conditional mask (x). The discriminator (D) then evaluates pairs—(G(x), x) for fake and (Y, x) for real—to drive the learning process.

2.3 Loss Function and Discriminator

The model employs an "Area Discriminator" that evaluates the image in blocks rather than as a single scalar. This forces the generator to maintain high fidelity in local details. The training is guided by a hybrid loss function combining Least Squares Loss (LSGAN) for stability and L1 regularization to minimize the pixel-wise difference between the generated target and the background.

3. Experimental Results

The model was evaluated using a dataset of 420 high-resolution sea surface image pairs ($256 \times 256$ pixels). The proposed CGAN method was compared against two baseline autoencoder models with varying residual blocks.

Model PSNR SSIM
Proposed CGAN Method 18.512 88.47%
Autoencoder (9 residual blocks) 17.596 81.65%
Autoencoder (6 residual blocks) 17.089 76.31%

The results demonstrate that the CGAN approach produces sharper images with higher structural consistency compared to the baselines.

4. Conclusion

This research presents an effective method for generating controllable and diverse remote sensing target images. By combining U-Net architectures with conditional masks, the model successfully addresses the data scarcity problem in satellite imagery analysis. Future work will explore applying this method to more complex land-based targets and larger-scale datasets.

Summary updated

Abstract

Abstract In view of the uncontrollable process of traditional GAN generating remote sensing target images, the shortcomings of generated samples are similar, and lack diversity. This paper proposes a generative confrontation network model based on background conditions. First, the computer vision attention mechanism is introduced into the generative confrontation network. Choose a learning target model so that the GAN network only learns the target model during training, and ignores other non-target information. Reduce the dependence on the number of samples in the GAN training process; secondly, use the U-net network as a generator to restore other non-target information when generating the remote sensing image of the target as much as possible; again, distinguish by different colors of the conditional mask The category of the generated target; at the same time, the L1 regularization loss is added to the loss term of the generator model, and finally, the remote sensing target image is generated. The experimental results show that the peak signal-to-noise ratio (PSNR) of the remote sensing image generation algorithm proposed in this paper reached 18.512, and the structural similarity (SSIM) reached 88.47%, which is better than the comparison test model where the generator is an ordinary autoencoder.
BibTeX
@article{liu2020conditional,
  author = {Haoyang Liu and Zhiyi Hu and Jun Yu and Shouyi Gao},
  title = {{Conditional GAN-based Remote Sensing Target Image Generation Method}},
  journal = {International Journal of Advanced Network Monitoring and Controls},
  year = {2020},
  doi = {10.21307/ijanmc-2020-039},
}
Figure