Search

The availability of high-resolution satellite images has accelerated the creation of new datasets designed to tackle broader remote sensing (RS) problems. Although popular tasks, such as scene classification, have received significant attention, the recent release of the Land-1.0 RS dataset marks the initiation of endeavors to estimate land-use and land-cover (LULC) fraction values per RGB satellite image. This challenging problem involves estimating LULC composition, i.e., the proportion of different LULC classes from satellite imagery, with major applications in environmental monitoring, agricultural/urban planning, and climate change studies. Currently, supervised deep learning models—the state-of-the-art in image classification—require large volumes of labeled training data to provide good generalization. To face the challenges posed by the scarcity of labeled RS data, self-supervised learning (SSL) models have recently emerged, learning directly from unlabeled data by leveraging the underlying structure. This is the first article to investigate the performance of SSL in LULC fraction estimation on RGB satellite patches using in-domain knowledge. We also performed a complementary analysis on LULC scene classification. Specifically, we pretrained Barlow Twins, MoCov2, SimCLR, and SimSiam SSL models with ResNet-18 using the Sentinel2GlobalLULC small RS dataset and then performed transfer learning to downstream tasks on Land-1.0. Our experiments demonstrate that SSL achieves competitive or slightly better results when trained on a smaller high-quality in-domain dataset of 194 877 samples compared to the supervised model trained on ImageNet-1k with 1 281 167 samples. This outcome highlights the effectiveness of SSL using in-distribution datasets, demonstrating efficient learning with fewer but more relevant data.

To test the hypothesis that a higher tendency to spontaneously focus on multiplicative relations (SFOR) leads to improvements in rational number knowledge via more exact estimation of fractional quantities, we presented sixth graders (n = 112) with fraction number line estimations and a novel task in which numerical information embedded in narratives could be estimated as fractions. Consistent with our main hypothesis, we found that SFOR tendency predicted both forms of fraction estimation. However, the relation between SFOR and fraction magnitude comparisons was mediated by fraction estimation, both on the number line and on whole number relations embedded in narrative vignettes. Thus, a higher tendency to recognize multiplicative relations in non-explicitly mathematical situations may contribute to increases in the precision with which students encode fractional relations in everyday contexts, both as spatial and numerical relations. [ABSTRACT FROM AUTHOR]

Recent years have seen a boom in research on the magnitude representations of common fractions and decimal fractions. However, previous studies generally only explored the magnitude representations of these two types of fractions in isolation. The present study, using a pretest-training-posttest design, examined and compared the magnitude representations of both types of fractions, and more importantly examined their relations to fraction translation and overall mathematical competence. The results showed that (1) accurate magnitude representations developed more often for decimal fractions than for common fractions; (2) individual differences in fraction magnitude representations predicted individual differences in fraction translation performance above and beyond initial knowledge of fraction translation; (3) mere exposure to visual representations of fractions on number lines improved children's mental representations of fraction magnitudes more than exposure to visual representations of fractions on circles or reading a textbook section on fractions. The present findings suggest that, like whole numbers, both types of fractions are organized on a mental number line, which appears to provide a central conceptual structure for organizing a wide range of numerical knowledge. Educational implications of the findings are discussed. [ABSTRACT FROM AUTHOR]