Channel-Attention-Based DenseNet Network for Remote Sensing Image Scene Classification

Remote sensing image scene classification has been widely applied and has attracted increasing attention. Recently, convolutional neural networks (CNNs) have achieved remarkable results in scene classification. However, scene images have complex semantic relationships between multiscale ground objects, and the traditional stacked network structure lacks the ability to effectively extract multiscale and key features, resulting in limited feature representation capabilities. By simulating the way that humans understand and perceive images, attention mechanisms can be beneficial for quickly and accurately acquiring key features. In our study, we propose a channel-attention-based DenseNet (CAD) network for scene classification. First, the lightweight DenseNet121 is selected as the backbone for the spatial relationship between multiscale ground objects. In the spatial domain, densely connected CNN layers can extract spatial features at multiple scales and correlate with each other. Second, in the channel domain, a channel attention mechanism is introduced to strengthen the weights of the important feature channels adaptively and to suppress the secondary feature channels. Third, the cross-entropy loss function based on label smoothing is used to reduce the impact of interclass similarity upon feature representations. The proposed CAD network is evaluated on three public datasets. The experimental results demonstrate that the CAD network can achieve performance comparable to those of other state-of-the-art methods. The visualization through the Grad-CAM ++ algorithm also reflects the effectiveness of channel attention and the powerful feature representation capabilities of the CAD network.