Air pollution is a serious problem in the world, and it is necessary to monitor air pollution emission sources in other neighboring countries to respond to the problem of air pollution spreading across borders. In this study, we utilized domestic and international optical images from KOMPSAT-3/3A satellites to build an AI training dataset for classifying industrial parks and quarries, which are representative sources of air pollution emissions. The data can be used to identify the distribution of air pollution emission sources located at home and abroad along with various state-of-the-art models in the image segmentation field, and is expected to contribute to the preservation of Korea’s air environment as a basis for establishing air-related policies.

Artificial intelligence (AI) has become the mainstream of analysis tools in remote sensing. Various semantic segmentation models have been introduced to segment land cover from aerial or satellite images, and remarkable results have been achieved. However, they often lack universal performance on unseen images, making them challenging to provide as a service. One of the primary reasons for the lack of robustness is overfitting, resulting from errors and inconsistencies in the ground truth (GT). In this study, we propose a method to mitigate these inconsistencies by utilizing redundant models and verify the improvement using a public dataset based on Google Earth images. Redundant models share the same network architecture and hyperparameters but are trained with different combinations of training and validation data on the same dataset. Because of the variations in sample exposure during training, these models yield slightly different inference results. This variability allows for the estimation of pixel-level confidence levels for the GT. The confidence level is incorporated into the GT to influence the loss calculation during the training of the enhanced model. Furthermore, we implemented a consensus model that employs modified masks, where classes with low confidence are substituted by the dominant classes identified through a majority vote from the redundant models. To further improve robustness, we extended the same approach to fuse the dataset with different class compositions based on imagery from the Korea Multipurpose Satellite 3A (KOMPSAT-3A). Performance evaluations were conducted on three network architectures: a simple network, U-Net, and DeepLabV3. In the single-dataset case, the performance of the enhanced and consensus models improved by an average of 2.49% and 2.59% across the network architectures. In the multi-dataset scenario, the enhanced models and consensus models showed an average performance improvement of 3.37% and 3.02% across the network architectures, respectively, compared to an average increase of 1.55% without the proposed method.

In South Korea, rural areas have been recognized for their potential as sustainable spaces for the future, but they are currently facing major problems. Unplanned construction of facilities such as factories, livestock facilities, and solar panels near residential areas is destroying the rural environment and deteriorating the quality of life of residents. Detection and monitoring of rural facilities are necessary to prevent disorderly development in rural areas and to manage rural space in a planned manner. In this study, satellite imagery data was utilized to obtain information on rural areas, which is useful for observing large areas and monitoring time series changes compared to field surveys. In this study, KOMPSAT ortho-mosaic optical imagery from 2019 and 2020 were utilized to construct AI training datasets for rural hazardous facilities segmentation for Seosan, Anseong, Naju, and Geochang areas. The dataset can be used in image segmentation models to classify rural facilities and can be used to monitor potentially hazardous facilities in rural areas. It is expected to contribute to solving rural problems by serving as the basis for rural planning.

Jae Young Chang, Kwan-Young Oh, Sun-Gu Lee. Korean J. Remote Sens. 2024;40:1397-408. https://doi.org/10.7780/kjrs.2024.40.6.3.2

Remote sensing images are often degraded by atmospheric effects, low illumination, and off-nadir viewing, which reduces the segmentation performance of deep models. KOMPSAT (Korea Multi-Purpose Satellite) imagery suffers from quality degradation because the Korean Peninsula is surrounded by sea on three sides and is subject to frequent weather and atmospheric variations. In practice, operators apply heuristic image enhancement techniques by hand, but these approaches are labor-intensive and inconsistent. To address this issue, we have proposed Adaptive Remote Sensing Image Enhancement (ARSIE), an automated reinforcement learning–based framework that improves segmentation performance on degraded KOMPSAT imagery. ARSIE takes only an existing segmentation network and training data as input, and learns, for each image, a sequence of enhancement operations selected from a filter pool. The policy network uses intermediate feature maps from the segmentation model to choose the next operation, ensuring that enhancement decisions directly support downstream segmentation performance. Experimental results show that ARSIE automatically discovers image-specific enhancement combinations and consistently improves segmentation accuracy on degraded KOMPSAT imagery. We demonstrate that ARSIE has the potential to be extended to improving the quality of other satellite imagery.