FADConv: A frequency-aware dynamic convolution for cropland non-agriculturalization identification and segmentation

Cropland non-agriculturalization refers to the conversion of arable land into non-agricultural uses such as forests, residential areas, and construction sites. This phenomenon not only directly leads to the loss of cropland resources but also poses systemic threats to food security and agricultural sustainability. Accurate identification of cropland and non-cropland areas is crucial for detecting and addressing this issue. Although remote sensing and deep learning methods have shown promise in cropland segmentation, challenges persist in misidentification and omission errors, particularly with high-resolution remote sensing imagery. Traditional CNNs employ static convolution layers, while dynamic convolution studies demonstrate that adaptively weighting multiple convolutional kernels through attention mechanisms can enhance accuracy. However, existing dynamic convolution methods relying on Global Average Pooling (GAP) for attention weight allocation suffer from information loss, limiting segmentation precision. This paper proposes Frequency-Aware Dynamic Convolution (FADConv) and a Frequency Attention (FAT) module to address these limitations. Building upon the foundational structure of dynamic convolution, we designed FADConv by integrating 2D Discrete Cosine Transform (2D DCT) to capture frequency domain features and fuse them. The FAT module generates high-quality attention weights that replace the traditional GAP method, making the combination between dynamic convolution kernels more reasonable. Experiments on the GID and Hi-CNA datasets demonstrate that FADConv significantly improves segmentation accuracy with minimal computational overhead. For instance, ResNet18 with FADConv achieves 1.9% and 2.7% increases in F1-score and IoU for cropland segmentation on GID, with only 58.87M additional MAdds. Compared to other dynamic convolution approaches, FADConv exhibits superior performance in cropland segmentation tasks. The code is obtained from this link: https://github.com/er-go-proxy/FADConv . • FADConv is committed to addressing the issue of non-agriculturalization. • FADConv integrates frequency features with dynamic convolution. • The FAT module uses DCT to improve the attention mechanism. • +2.7% IoU gain on GID dataset with minimal computational overhead. • FAT generalizes to other dynamic convolutions (e.g., ODConv, KW).