CloudformerV3: Multi-Scale Adapter and Multi-Level Large Window Attention for Cloud Detection

Abstract

Cloud detection in remote sensing images is a crucial preprocessing step that efficiently identifies and extracts cloud-covered areas within the images, ensuring the precision and reliability of subsequent analyses and applications. Given the diversity of clouds and the intricacies of the surface, distinguishing the boundaries between thin clouds and the underlying surface is a major challenge in cloud detection. To address these challenges, an advanced cloud detection method, CloudformerV3, is presented in this paper. The proposed method employs a multi-scale adapter to incorporate dark and bright channel prior information into the model’s backbone, enhancing the model’s ability to capture prior information and multi-scale details from remote sensing images. Additionally, multi-level large window attention is utilized, enabling high-resolution feature maps and low-resolution feature maps to mutually focus and subsequently merge during the resolution recovery phase. This facilitates the establishment of connections between different levels of feature maps and offers comprehensive contextual information for the model’s decoder. The experimental results on the GF1_WHU dataset illustrate that the method proposed in this paper achieves MIoU of 92.89%, while achieving higher detection accuracy compared to state-of-the-art cloud detection models. Specifically, in comparison to Cloudformer, our method demonstrates a 1.11% improvement, while compared to CloudformerV2, there is a 0.37% increase. Furthermore, enhanced detection performance is achieved along cloud edges and concerning thin clouds, showcasing the efficacy of the proposed method.