With the continuing improvement of remote-sensing (RS) sensors, it is crucial to monitor Earth surface changes at fine scale and in great detail. Thus, semantic change detection (SCD), which is capable of locating and identifying “from-to” change information simultaneously, is gaining growing attention in RS community. However, due to the limitation of large-scale SCD datasets, most existing SCD methods are focused on scene-level changes, where semantic change maps are generated with only coarse boundary or scarce category information. To address this issue, we propose a novel convolutional network for large-scale SCD (SCDNet). It is based on a Siamese UNet architecture, which consists of two encoders and two decoders with shared weights. First, multi-temporal images are given as input to the encoders to extract multi-scale deep representations. A multi-scale atrous convolution (MAC) unit is inserted at the end of the encoders to enlarge the receptive field as well as capturing multi-scale information. Then, difference feature maps are generated for each scale, which are combined with feature maps from the encoders to serve as inputs for the decoders. Attention mechanism and deep supervision strategy are further introduced to improve network performance. Finally, we utilize softmax layer to produce a semantic change map for each time image. Extensive experiments are carried out on two large-scale high-resolution SCD datasets, which demonstrates the effectiveness and superiority of the proposed method.

随着遥感 (RS) 传感器的不断改进，以精细的尺度和非常详细地监测地球表面变化至关重要。因此，能够同时定位和识别“从到”变化信息的语义变化检测（SCD）在 RS 社区中越来越受到关注。然而，由于大规模 SCD 数据集的限制，大多数现有的 SCD 方法都集中在场景级别的变化上，其中生成的语义变化图仅具有粗边界或稀缺类别信息。为了解决这个问题，我们提出了一种用于大规模 SCD 的新型卷积网络（SCDNet）。它基于 Siamese UNet 架构，由两个编码器和两个具有共享权重的解码器组成。首先，将多时间图像作为输入到编码器以提取多尺度深度表示。在编码器的末端插入多尺度多孔卷积 (MAC) 单元以扩大感受野并捕获多尺度信息。然后，为每个尺度生成差异特征图，这些特征图与来自编码器的特征图结合作为解码器的输入。进一步引入注意力机制和深度监督策略以提高网络性能。最后，我们利用 softmax 层为每个时间图像生成语义变化图。在两个大规模高分辨率 SCD 数据集上进行了大量实验，证明了所提出方法的有效性和优越性。为每个尺度生成差异特征图，这些特征图与来自编码器的特征图结合作为解码器的输入。进一步引入注意力机制和深度监督策略以提高网络性能。最后，我们利用 softmax 层为每个时间图像生成语义变化图。在两个大规模高分辨率 SCD 数据集上进行了大量实验，证明了所提出方法的有效性和优越性。为每个尺度生成差异特征图，这些特征图与来自编码器的特征图结合作为解码器的输入。进一步引入注意力机制和深度监督策略以提高网络性能。最后，我们利用 softmax 层为每个时间图像生成语义变化图。在两个大规模高分辨率 SCD 数据集上进行了大量实验，证明了所提出方法的有效性和优越性。