Weak Labeling for Cropland Mapping in Africa
This addresses a scalability bottleneck in cropland mapping for agricultural and food security applications in Africa, but it is incremental as it builds on existing weak labeling methods.
The paper tackles the problem of limited high-resolution cropland maps in Africa by proposing an approach that refines weak labels using unsupervised object clustering, combined with sparse human annotations, to train a semantic segmentation network. The result shows an increase in F1 score for cropland from 0.53 to 0.84 when adding mined negative labels with only 33 human annotations.
Cropland mapping can play a vital role in addressing environmental, agricultural, and food security challenges. However, in the context of Africa, practical applications are often hindered by the limited availability of high-resolution cropland maps. Such maps typically require extensive human labeling, thereby creating a scalability bottleneck. To address this, we propose an approach that utilizes unsupervised object clustering to refine existing weak labels, such as those obtained from global cropland maps. The refined labels, in conjunction with sparse human annotations, serve as training data for a semantic segmentation network designed to identify cropland areas. We conduct experiments to demonstrate the benefits of the improved weak labels generated by our method. In a scenario where we train our model with only 33 human-annotated labels, the F_1 score for the cropland category increases from 0.53 to 0.84 when we add the mined negative labels.