Multi-class geospatial object detection in high-resolution remote sensing images has significant potential in various domains such as industrial production, military warning, disaster monitoring, and urban planning. However, the traditional process of remote sensing object detection involves several time-consuming steps, including image acquisition, image download, ground processing, and object detection. These steps may not be suitable for tasks with shorter timeliness requirements, such as military warning and disaster monitoring. Additionally, the transmission of massive data from satellites to the ground is limited by bandwidth, resulting in time delays and redundant information, such as cloud coverage images. To address these challenges and achieve efficient utilization of information, this paper proposes a comprehensive on-board multi-class geospatial object detection scheme. The proposed scheme consists of several steps. Firstly, the satellite imagery is sliced, and the PID-Net (Proportional-Integral-Derivative Network) method is employed to detect and filter out cloud-covered tiles. Subsequently, our Manhattan Intersection over Union (MIOU) loss-based YOLO (You Only Look Once) v7-Tiny method is used to detect remote-sensing objects in the remaining tiles. Finally, the detection results are mapped back to the original image, and the truncated NMS (Non-Maximum Suppression) method is utilized to filter out repeated and noisy boxes. To validate the reliability of the scheme, this paper creates a new dataset called DOTA-CD (Dataset for Object Detection in Aerial Images-Cloud Detection). Experiments were conducted on both ground and on-board equipment using the AIR-CD dataset, DOTA dataset, and DOTA-CD dataset. The results demonstrate the effectiveness of our method.

Aircraft detection in SAR images of airports remains crucial for continuous ground observation and aviation transportation scheduling in all weather conditions, but low resolution and complex scenes pose unique challenges. Existing methods struggle with accuracy, overlapping detections, and missed targets. We propose You Only Look Once-SAR Aircraft Detector (YOLO-SAD), a novel detector that tackles these issues. YOLO-SAD leverages the Attention-Efficient Layer Aggregation Network-Head (A-ELAN-H) module to prioritize essential features for improved accuracy. Additionally, the SAR Aircraft Detection-Feature Pyramid Network (SAD-FPN) optimizes multi-scale feature fusion, boosting detection speed. Finally, Enhanced Non-Maximum Suppression (EH-NMS) eliminates overlapping detections. On the SAR Aircraft Detection Dataset (SADD), YOLO-SAD achieved 91.9% AP(0.5) and 57.1% AP(0.5:0.95), surpassing the baseline by 2.1% and 1.9%, respectively. Extensive comparisons on SADD further demonstrate YOLO-SAD’s superiority over five state-of-the-art methods in both AP(0.5) and AP(0.5:0.95). The outcomes of further comparative experiments on the SAR-AIRcraft-1.0 dataset confirm the robust generalization capability of YOLO-SAD, demonstrating its potential use in aircraft detection with SAR.

One recent trend in optical remote sensing is to increase observation frequencies. However, there are still challenges on the night side when sunlight is not available. Due to their powerful capabilities in low-light sensing, nightlight satellite sensors have been deployed to capture nightscapes of Earth from space, observing anthropomorphic and natural activities at night. To date, the mainstream of nightlight remote sensing applications has mainly focused on artificial lights, especially within cities or self-luminous bodies, such as fisheries, oil, offshore rigs, etc. Observations taken under moonlight are often discarded or corrected to reduce lunar effects. Some researchers have discussed the possibility of using moonlight as a useful illuminating source at night for the detection of nocturnal features on Earth, but no quantitative analysis has been reported so far. This study aims to systematically evaluate the potential of moonlight remote sensing with mono-spectral Visible Infrared Imaging Radiometer Suite/Day-Night-Band (VIIRS/DNB) imagery and multi-spectral photos taken by astronauts from the International Space Station (ISS), as well as unmanned aerial vehicle (UAV) night-time imagery. Using the VIIRS/DNB, ISS and UAV moonlight images, the possibilities of the moonlight remote sensing were first discussed. Then, the VIIRS/DNB, ISS, UAV images were classified over different non-self-lighting land surfaces to explore the potential of moonlight remote sensing. The overall accuracies (OA) and kappa coefficients are 79.80% and 0.45, 87.16% and 0.77, 91.49% and 0.85, respectively, indicating a capability to characterize land surface that is very similar to daytime remote sensing. Finally, the characteristics of current moonlight remote sensing are discussed in terms of bands, spatial resolutions, and sensors. The results confirm that moonlight remote sensing has huge potential for Earth observation, which will be of great importance to significantly increase the temporal coverage of optical remote sensing during the whole diurnal cycle. Based on these discussions, we further examined requirements for next-generation nightlight remote sensing satellite sensors.

Accurate monitoring of the spatiotemporal dynamics of snow and ice is essential for under-standing and predicting the impacts of climate change on Arctic ecosystems and their feedback on global climate. Traditional optical and Synthetic Aperture Radar (SAR) remote sensing still have limitations in the long-time series observation of polar regions. Although several studies have demonstrated the potential of moonlight remote sensing for mapping polar snow/ice covers, systematic evaluation on applying moonlight remote sensing to monitoring spatiotemporal dynamics of polar snow/ice covers, especially during polar night periods is highly demanded. Here we present a systematic assessment in Svalbard, Norway and using data taken from the Visible Infrared Imaging Radiometer Suite (VIIRS) on the Suomi National Polar-orbiting Partnership (SNPP) Day/Night Band (DNB) sensor to monitor the spatiotemporal dynamics of snow/ice covers during dark Arctic winters when no solar illumination available for months. We successfully revealed the spatiotemporal dynamics of snow/ice covers from 2012 to 2022 during polar night/winter periods, using the VIIRS/DNB time series data and the object-oriented Random Forests (RF) algorithm, achieving the average accuracy and kappa coefficient of 96.27% and 0.93, respectively. Our findings indicate that the polar snow/ice covers show seasonal and inter-seasonal dynamics, thus requiring more frequent observations. Our results confirm and realize the potential of moonlight remote sensing for continuous monitoring of snow/ice in the Arctic region and together with other types of remote sensing data, moonlight remote sensing will be a very useful tool for polar studies and climate change.

Efficient multi-object tracking (MOT) in satellite videos is crucial for numerous applications, ranging from surveillance to environmental monitoring. Existing methods often struggle with effectively exploring the correlation and contextual cues inherent in the consecutive features of video sequences, resulting in redundant feature inference and unreliable motion estimation for tracking. To address these challenges, we propose the MFACNet, a novel multi-frame features aggregating and inter-feature correlation framework for enhancing MOT in satellite videos with the idea of utilizing the features of consecutive frames. The MFACNet integrates multi-frame feature aggregation techniques with inter-feature correlation mechanisms to improve tracking accuracy and robustness. Specifically, our framework leverages temporal information across the features of consecutive frames to capture contextual cues and refine object representations over time. Moreover, we introduce a mechanism to explicitly model the correlations between adjacent features in video sequences, facilitating a more accurate motion estimation and trajectory associations. We evaluated the MFACNet using benchmark datasets for satellite-based video MOT tasks and demonstrated its superiority in terms of tracking accuracy and robustness over state-of-the-art performance by 2.0% in MOTA and 1.6% in IDF1. Our experimental results highlight the potential of precisely utilizing deep features from video sequences.