Semi-Supervised Remote Sensing Imagery Scene Classification Based on Probabilistic Selection

To address low-quality pseudo-labels and class imbalance in semi-supervised remote sensing scene classification, this letter proposes a novel dynamic adaptive reweighting and pseudo-labeling method (DARP) based on probabilistic random selection. First, we present a semi-supervised recursive learning strategy with adaptive category thresholds and probabilistic random selection approach to solve the problems of sample diversity and imbalanced distribution of pseudo-labeled sample categories. Second, we design an image category weight determination method based on the number and accuracy of image categories and assign larger weights to classes with lower accuracy, thereby increasing the probability of these class samples being selected and improving the overall accuracy. Experiments were conducted on the publicly available datasets UCM and NWPU-RESISC45, and the accuracies reached 95.87% and 88.86%, using only five labeled samples per category, which is an improvement of 5.71% and 1.56% over existing semi-supervised methods.