arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.07428eess.IV

BiGSeT：用于基于 DNN 的高光谱异常检测的二值掩码引导分离训练

BiGSeT: Binary Mask-Guided Separation Training for DNN-based Hyperspectral Anomaly Detection

Haijun Liu, Xi Su, Xiangfei Shen, Lihui Chen, Xichuan Zhou

备注：
13 pages, 13 figures, submitted to IEEE TRANSACTIONS ON IMAGE PROCESSING
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对 DNN 高光谱异常检测中的恒等映射问题，提出 BiGSeT 二值掩码分离训练策略，利用 LoG 抑制异常并提升多种网络的检测性能。

摘要：

高光谱异常检测（HAD）旨在在没有先验知识的情况下，识别出少数在光谱上与其周围背景不同的异常。深度神经网络（DNN），包括自动编码器（AE）、卷积神经网络（CNN）和视觉 Transformer（ViT），由于其对复杂背景建模的强大能力，已在该领域表现出卓越性能。然而，对于重建任务，DNN 往往会将背景和异常同时纳入估计背景中，这被称为恒等映射问题（IMP），并导致性能显著下降。为解决这一局限，我们提出了一种与模型无关、由二值掩码引导的 DNN 分离训练策略，名为 BiGSeT。我们的方法引入了一种基于潜在二值掩码的分离训练损失，以分别约束估计图像中的背景和异常。背景得到保留，同时利用高效的二阶高斯-拉普拉斯（LoG）算子抑制潜在异常，从而生成纯净的背景估计。为在训练过程中保持可分离性，我们使用训练前估计的鲁棒比例阈值周期性更新掩码。在实验中，我们采用一个普通 AE 作为网络，在若干真实世界数据集上验证该训练策略。结果表明，与一些最先进方法相比，其性能更优。具体而言，我们在 HyMap Cooke City 数据集上取得了 90.67% 的 AUC 分数。此外，我们将该训练策略应用于其他深度网络结构，与原始版本相比均提升了检测性能，证明了其有效的可迁移性。我们方法的代码将在 https://github.com/enter-i-username/BiGSeT 提供。

英文摘要：

Hyperspectral anomaly detection (HAD) aims to recognize a minority of anomalies that are spectrally different from their surrounding background without prior knowledge. Deep neural networks (DNNs), including autoencoders (AEs), convolutional neural networks (CNNs) and vision transformers (ViTs), have shown remarkable performance in this field due to their powerful ability to model the complicated background. However, for reconstruction tasks, DNNs tend to incorporate both background and anomalies into the estimated background, which is referred to as the identical mapping problem (IMP) and leads to significantly decreased performance. To address this limitation, we propose a model-independent binary mask-guided separation training strategy for DNNs, named BiGSeT. Our method introduces a separation training loss based on a latent binary mask to separately constrain the background and anomalies in the estimated image. The background is preserved, while the potential anomalies are suppressed by using an efficient second-order Laplacian of Gaussian (LoG) operator, generating a pure background estimate. In order to maintain separability during training, we periodically update the mask using a robust proportion threshold estimated before the training. In our experiments, We adopt a vanilla AE as the network to validate our training strategy on several real-world datasets. Our results show superior performance compared to some state-of-the-art methods. Specifically, we achieved a 90.67% AUC score on the HyMap Cooke City dataset. Additionally, we applied our training strategy to other deep network structures, achieving improved detection performance compared to their original versions, demonstrating its effective transferability. The code of our method will be available at https://github.com/enter-i-username/BiGSeT.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑