NFANet: A Novel Method for Weakly Supervised Water Extraction from High-Resolution Remote Sensing Imagery
This addresses the difficulty of obtaining precise pixel-level labels for remote sensing applications, though it appears incremental as it builds on existing weakly supervised techniques.
The paper tackles the problem of water extraction from high-resolution remote sensing imagery by proposing NFANet, a weakly supervised method that uses point labels instead of pixel-level labels, achieving results comparable to state-of-the-art approaches.
The use of deep learning for water extraction requires precise pixel-level labels. However, it is very difficult to label high-resolution remote sensing images at the pixel level. Therefore, we study how to utilize point labels to extract water bodies and propose a novel method called the neighbor feature aggregation network (NFANet). Compared with pixellevel labels, point labels are much easier to obtain, but they will lose much information. In this paper, we take advantage of the similarity between the adjacent pixels of a local water-body, and propose a neighbor sampler to resample remote sensing images. Then, the sampled images are sent to the network for feature aggregation. In addition, we use an improved recursive training algorithm to further improve the extraction accuracy, making the water boundary more natural. Furthermore, our method utilizes neighboring features instead of global or local features to learn more representative features. The experimental results show that the proposed NFANet method not only outperforms other studied weakly supervised approaches, but also obtains similar results as the state-of-the-art ones.