Abstract In a structural health monitoring (SHM) system that uses digital cameras to monitor cracks of structural surfaces, techniques for reliable and effective data compression are essential to ensure a stable and energy-efficient crack images transmission in wireless devices, e.g., drones and robots with high definition cameras installed. Compressive sensing (CS) is a signal processing technique that allows accurate recovery of a signal from a sampling rate much smaller than the limitation of the Nyquist sampling theorem. Different from the popular approach of simultaneously training encoder and decoder using neural network models, the CS theory ensures a high probability of accurate signal reconstruction based on random measurements that is shorter than the length of the original signal under a sparsity constraint. Such method is particularly useful when measurements are expensive, such as wireless sensing of civil structures, because its hardware implementation allows down sampling of signals during the sensing process. Hence, CS methods can achieve significant energy saving for the sensing devices. However, the strong assumption of the signals being highly sparse in an invertible space is relatively hard to guarantee for many real images, such as image of cracks. In this paper, we present a new approach of CS that replaces the sparsity regularization with a generative model that is able to effectively capture a low dimension representation of targeted images. We develop a recovery framework for automatic crack segmentation of compressed crack images based on this new CS method. We demonstrate the remarkable performance of our method that takes advantage of the strong capability of generative models to capture the necessary features required in the crack segmentation task even the backgrounds of the generated images are not well reconstructed. The superior performance of our recovery framework is illustrated by comparisons to three existing CS algorithms. Furthermore, we show that our framework is potentially extensible to other common problems in automatic crack segmentation, such as defect recovery from motion blurring and occlusion.

中文翻译：

---

使用生成模型恢复用于自动裂缝分割的压缩图像

摘要 在使用数码相机监测结构表面裂纹的结构健康监测 (SHM) 系统中，可靠和有效的数据压缩技术对于确保无线设备（例如无人机和机器人）中稳定且节能的裂纹图像传输至关重要。安装了高清摄像头。压缩感知 (CS) 是一种信号处理技术，它允许从远小于奈奎斯特采样定理限制的采样率中准确恢复信号。与使用神经网络模型同时训练编码器和解码器的流行方法不同，CS 理论确保基于随机测量的准确信号重建的高概率，该随机测量比稀疏约束下原始信号的长度短。这种方法在测量成本高昂时特别有用，例如土木结构的无线传感，因为它的硬件实现允许在传感过程中对信号进行下采样。因此，CS方法可以实现传感设备的显着节能。然而，对于许多真实图像，例如裂缝图像，信号在可逆空间中高度稀疏的强假设相对难以保证。在本文中，我们提出了一种新的 CS 方法，用能够有效捕获目标图像的低维表示的生成模型代替稀疏正则化。我们基于这种新的 CS 方法开发了一个用于压缩裂纹图像自动裂纹分割的恢复框架。我们展示了我们的方法的卓越性能，该方法利用生成模型的强大能力来捕获裂缝分割任务所需的必要特征，即使生成的图像的背景没有得到很好的重建。我们的恢复框架的卓越性能通过与三种现有 CS 算法的比较来说明。此外，我们表明我们的框架有可能扩展到自动裂缝分割中的其他常见问题，例如从运动模糊和遮挡中恢复缺陷。我们的恢复框架的卓越性能通过与三种现有 CS 算法的比较来说明。此外，我们表明我们的框架有可能扩展到自动裂缝分割中的其他常见问题，例如从运动模糊和遮挡中恢复缺陷。我们的恢复框架的卓越性能通过与三种现有 CS 算法的比较来说明。此外，我们表明我们的框架有可能扩展到自动裂缝分割中的其他常见问题，例如从运动模糊和遮挡中恢复缺陷。