论文标题
使用双卷积神经网络中遥感图像中的多尺度云检测
Multi-scale Cloud Detection in Remote Sensing Images using a Dual Convolutional Neural Network
论文作者
论文摘要
卷积神经网络（CNN）的语义分割已在遥感图像的像素级分类中提高了最新技术。但是，处理大图像通常需要在小斑块中分析图像，因此具有较大空间范围的特征仍会在诸如云掩盖之类的任务中引起挑战。为了支持更广泛的空间特征，同时降低了大型卫星图像的计算要求，我们提出了两个级联的CNN模型组件的体系结构，连续处理了底漆和完整分辨率的图像。第一个组件将内部云区域的斑块与云边界区域的贴片区分开。对于需要进一步分割的云效能边缘贴片，该框架然后将计算委派给细颗粒模型组件。我们将体系结构应用于完整的Sentinel-2多光谱图像的云检测数据集，大约在土地使用应用程序中对最小的假否定性进行了注释。在此特定的任务和数据上，我们基于修补程序的CNN基线实现了16 \％的像素精度的相对提高。
Semantic segmentation by convolutional neural networks (CNN) has advanced the state of the art in pixel-level classification of remote sensing images. However, processing large images typically requires analyzing the image in small patches, and hence features that have large spatial extent still cause challenges in tasks such as cloud masking. To support a wider scale of spatial features while simultaneously reducing computational requirements for large satellite images, we propose an architecture of two cascaded CNN model components successively processing undersampled and full resolution images. The first component distinguishes between patches in the inner cloud area from patches at the cloud's boundary region. For the cloud-ambiguous edge patches requiring further segmentation, the framework then delegates computation to a fine-grained model component. We apply the architecture to a cloud detection dataset of complete Sentinel-2 multispectral images, approximately annotated for minimal false negatives in a land use application. On this specific task and data, we achieve a 16\% relative improvement in pixel accuracy over a CNN baseline based on patching.
