Many real-world applications rely on land surface temperature (LST) data at high spatiotemporal resolution. In complex urban areas, LST exhibits significant variations, fluctuating dramatically within and across city blocks. Landsat provides high spatial resolution data at 100 meters but is limited by long revisit time, with cloud cover further disrupting data collection. Here, we propose DELAG, a deep ensemble learning method that integrates annual temperature cycles and Gaussian processes, to reconstruct Landsat LST in complex urban areas. Leveraging the cross-track characteristics and dual-satellite operation of Landsat since 2021, we further enhance data availability to 4 scenes every 16 days. We select New York City, London and Hong Kong from three different continents as study areas. Experiments show that DELAG successfully reconstructed LST in the three cities under clear-sky (RMSE = 0.73-0.96 K) and heavily-cloudy (RMSE = 0.84-1.62 K) situations, superior to existing methods. Additionally, DELAG can quantify uncertainty that enhances LST reconstruction reliability. We further tested the reconstructed LST to estimate near-surface air temperature, achieving results (RMSE = 1.48-2.11 K) comparable to those derived from clear-sky LST (RMSE = 1.63-2.02 K). The results demonstrate the successful reconstruction through DELAG and highlight the broader applications of LST reconstruction for estimating accurate air temperature. Our study thus provides a novel and practical method for Landsat LST reconstruction, particularly suited for complex urban areas within Landsat cross-track areas, taking one step toward addressing complex climate events at high spatiotemporal resolution. Code and data will be available at skrisliu.com/delag.

Artificial light at night (ALAN) changes throughout the night and the year within an area. It is crucial to monitor the time-series characteristics of ALAN. In this study, we investigated the changes in ALAN in Hong Kong over time from two medium-resolution remote sensing products: Luojia 1-01 (LJ-1) and the International Space Station (ISS). We validated these images with in-situ night sky brightness measurements. The results show that both the calibrated LJ-1 and the uncalibrated ISS images achieved similar agreements with the in-situ data (R=0.73 and R=0.83, respectively). LJ-1 and ISS images are useful for evaluating ALAN changes over the years and throughout the night, respectively. In early nights (before 23:00), commercial areas were brighter, while in late nights (after 23:00), port facilities and airports were brighter. Finally, we highlight the importance of color composition, time-series observations within one night, and multi-angle observations within minutes for ALAN monitoring.

Fine-scale monitoring of particulate matter (PM) is essential to assess the patterns and sources of air pollution. However, to this date, its usage is still limited. In this study, we explore the usage of Landsat imagery for PM mapping at 30 m spatial resolution, where air pollutants at the street level can be distinguished. With in-situ data from the low-cost sensors of the PurpleAir network, we developed a multitask neural network that can simultaneously estimate all three metrics of PM, i.e., PM1, PM2.5, and PM10. With 936 monitoring stations from 25 dates, a total of 15,250 valid observations were used to construct the model. With external validation, results show that the model can achieve R2of 0.767, 0.803, and 0.799 in estimating PM1, PM2.5, and PM10, respectively, at the street level. Using one multitask neural network to predict PM1, PM2.5, and PM10 improves the efficiency of real-life air pollutant monitoring.

Remote sensing data provide a low-cost solution for large-scale monitoring of air pollution via the retrieval of aerosol optical depth (AOD), but is often limited by cloud contamination. Existing methods for AOD reconstruction rely on temporal information. However, for remote sensing data at high spatial resolution, multi-temporal observations are often unavailable. In this letter, we take advantage of deep representation learning from convolutional neural networks and propose Deep Feature Gaussian Processes (DFGP) for single-scene AOD reconstruction. By using deep learning, we transform the variables to a feature space with better explainable power. By using Gaussian processes, we explicitly consider the correlation between observed AOD and missing AOD in spatial and feature domains. Experiments on two AOD datasets with real-world cloud patterns showed that the proposed method outperformed deep CNN and random forest, achieving R2of 0.7431 on MODIS AOD and R2of 0.9211 on EMIT AOD, compared to deep CNN’s R2of 0.6507 and R2of 0.8619. The proposed methods increased R2by over 0.35 compared to the popular random forest in AOD reconstruction. The data and code used in this study are available at https://skrisliu.com/dfgp.

Earth Hour is an annual campaign to turn off non-essential electric lights for one hour as a symbol of commitment to our planet. Small gesture as it may be, this one-hour event may lead to temporary reduction of energy usage and light pollution. In Earth Hour 2021, we conducted a large survey in Hong Kong to see its impacts using multi-source open and volunteered-crowd-sourcing data. These data were collected from 120 traffic cameras, 27 weather cameras, 2 panoramic 360 cameras, 1 wide-field camera, 7 all-sky cameras, 10 cellphones of volunteers, and 1 camera recording videos onboard of tramway. By analyzing the multi-source images and videos, we identified 122 individual buildings that participated in Earth Hour 2021. We further demonstrate in detail using image analysis techniques the extent of light pollution reduction in an urban environment. Finally, we present a demo experiment to identify light sources using artificial intelligence (AI) techniques.

Although artificial light at night (ALAN) is essential for nighttime activities, any unregulated and abusive usage can lead to severe degradation in quality of life. In this study, we used high-resolution nighttime remote sensing data of 1-meter spatial resolution to investigate light sources in an urban area of Hong Kong with one million residents. We classified ALAN sources into three categories based on their origins: Building, Park, and Street. We found that 42% of light was from Building, a large fraction of which was unnecessary decorative lighting such as signboards. Lighting from Street accounted for 41%, whereas an unexpectedly high proportion was associated with Park (17%), with sport facilities-related lighting being the dominant contributor. We also detailed one case study which shows the disruptive effects of unregulated usage of LED signboards to the neighboring residential apartments.

Particulate matter (PM), such as PM2.5 and PM10, was the major pollutant in a severe air pollution episode in 2013 eastern China. Limited by the coverage of stations, fine-scale monitoring at every corner in the city is difficult, if not impossible. Hyperspectral imagery can capture the ground and air information, from which we can estimate the concentrations of PM. In this study, we develop a multitask learning method to estimate the concentrations of PM based on the 10-m hyperspectral data from the newly-launched Zhuhai-1 satellites. We first convert the raw radiance to top-of-atmosphere (TOA) reflectance using the 1985 Wehrli solar irradiance spectrum. Then, we train a multitask network to simultaneously estimate PM2.5 and PM10 concentrations based on the TOA hyperspectral data. Results show that our method leads to estimations of an R-squared of 0.77 for PM2.5 and an R-squared of 0.42 for PM10.

With the near completion of WUDAPT (World Urban Database and Access Portal Tools) Level 0 data, one of the next goals is to generate more accurate and detailed local climate zone (LCZ) maps. An important issue is how to integrate building height information into LCZ maps. We here present a multi-label classification method using very high resolution (VHR) imagery to implicitly integrate building height information. Since we humans can tell whether a place is high-rise or not based on the shading of buildings and the surrounding context, it is possible to extract such information using deep learning methods. We use Hong Kong as a case study and show the potential of LCZ mapping with VHR imagery in distinguishing small-scale landscape features like city parks. The multi-label LCZ maps also provide a solution to generate fine-grained subclass LCZ mapping, in which a place can be classified as a combination of multiple LCZs, e.g., compact low-rise with open high-rise.

Long-term monitoring of artificial light at night (ALAN) is essential for our understanding of the source of light pollution and developing mechanisms to control it. In this study, based on the VIIRS monthly product and land use data, we analyzed the long-term ALAN in Hong Kong between 2012 and 2019. We could not detect any long-term trend in the level of ALAN of Hong Kong from this dataset over the eight years of observations at the level of detection accuracy of the VIIRS monthly data. We performed a detailed analysis of the ALAN from Hong Kong and its relationship with land use classes. We found that in Hong Kong, the public residential areas are brighter than the private ones, likely the consequence of a combination of population density and lighting designs. Using the clustering method, we were able to identify some persistently bright (or dark) facilities, such as the Hong Kong-Zhuhai-Macau Bridge Port, airport and port facilities. Transient phenomena such as wildfires were identified as well. Finally, we observed a brighter background ALAN associated with an elevated humidity level$(\mathrm{R}=0.54)$, which can possibly be attributed to the dispersing effect of water vapor on radiation. Since large public transportation facilities emitted the most ALAN in Hong Kong, we suggest adopting sustainable design in future transportation projects to reduce the emitted ALAN to the space, thereby reducing light pollution.

Although deep learning has achieved great success in the image-classification tasks, its performance is subject to the quantity and quality of the training samples. For the classification of the polarimetric synthetic aperture radar (PolSAR) images, it is nearly impossible to annotate the images from visual interpretation. Therefore, it is urgent for remote-sensing scientists to develop new techniques for PolSAR image classification under the condition of very few training samples. In this letter, we take the advantage of active learning and propose active ensemble deep learning (AEDL) for PolSAR image classification. We first show that only 35% of the predicted labels of the deep-learning model's snapshots near its convergence were exactly the same. The disagreement between the snapshots is nonnegligible. From the perspective of multiview learning, the snapshots together serve as a good committee to evaluate the importance of the unlabeled instances. Using the snapshot committee to give out the informativeness of the unlabeled data, the proposed AEDL achieved better performance on two real PolSAR images than the standard active learning strategies. It achieved the same classification accuracy with only 86% and 55% of the training samples compared to the breaking tie active learning and random selection for the Flevoland data set.

Current hyperspectral image classification assumes that a predefined classification system is closed and complete, and there are no unknown or novel classes in the unseen data. However, this assumption may be too strict for the real world. Often, novel classes are overlooked when the classification system is constructed. The closed nature forces a model to assign a label given a new sample and may lead to overestimation of known land covers (e.g., crop area). To tackle this issue, we propose a multitask deep learning method that simultaneously conducts classification and reconstruction in the open world (named MDL4OW) where unknown classes may exist. The reconstructed data are compared with the original data; those failing to be reconstructed are considered unknown based on the assumption that they are not well represented in the latent features due to the lack of labels. A threshold needs to be defined to separate the unknown and known classes; we propose two strategies based on the extreme value theory for few- and many-shot scenarios. The proposed method was tested on real-world hyperspectral images; state-of-the-art results were achieved, e.g., improving the overall accuracy by 4.94% for the Salinas data. By considering the existence of unknown classes in the open world, our method achieved more accurate hyperspectral image classification, especially under the few-shot context.

In this letter, we propose a multitask deep learning method for the classification of multiple hyperspectral data in a single training. Deep learning models have achieved promising results on hyperspectral image classification, but their performance highly relies on sufficient labeled samples that are scarce on hyperspectral images. However, samples from multiple data sets might be sufficient to train one deep learning model, thereby improving its performance. To do so, we trained an identical feature extractor for all data, and the extracted features were fed into corresponding softmax classifiers. Spectral knowledge was introduced to ensure that the shared features were similar across domains. Four hyperspectral data sets were used in the experiments. We achieved higher classification accuracies on three data sets (Pavia University, Pavia Center, and Indian Pines) and competitive results on the Salinas Valley data compared with the baseline. Spectral knowledge was useful to prevent the deep network from overfitting when the data shared similar spectral response. The proposed method tested on two deep CNNs successfully shows its ability to utilize samples from multiple data sets and to enhance networks' performance.

In this paper, we propose a wide contextual residual network (WCRN) with active learning (AL) for remote sensing image (RSI) classification. Although ResNets have achieved great success in various applications (e.g. RSI classification), its performance is limited by the requirement of abundant labeled samples. As it is very difficult and expensive to obtain class labels in real world, we integrate the proposed WCRN with AL to improve its generalization by using the most informative training samples. Specifically, we first design a wide contextual residual network for RSI classification. We then integrate it with AL to achieve good machine generalization with limited number of training sampling. Experimental results on the University of Pavia and Flevoland datasets demonstrate that the proposed WCRN with AL can significantly reduce the needs of samples.