In order to effectively manage agriculture and promote sustainable land use in Africa, accurate mapping of irrigated croplands is imperative. This paper pioneers a novel Hybrid Vegetation Index (HVI) and integrated an approach harnessing the synergies between Radar imagery from Sentinel-1 and optical data from Sentinel-2, key vegetation indices (HVI, MSAVI), and Principal Component Analysis (PCA) to significantly enhance mapping accuracy across diverse African landscapes. Employing Convolutional Neural Network (CNN), Random Forest (RF), and Support Vector Machine (SVM) classifiers, the combined potential of these satellite datasets is thoroughly Assessed. The integrated utilization of both Sentinel-1 and Sentinel-2 demonstrates a substantial enhancement in overall accuracy, elevating classification results by approximately 5-6% when compared to individual sensor applications. Specifically, in Kenya, Egypt, and Nigeria, the amalgamation of Sentinel-1 and Sentinel-2 enables CNN to achieve remarkable accuracies of 97.5%, 98.01%, and 98%, respectively. These findings underscore the superior performance of the fused Sentinel data and highlighted the pivotal role of their integration in advancing precise agricultural mapping. This research emphasizes the promising impact of HVI and Sentinel satellite missions in empowering informed decision-making and promoting sustainable land management practices across vital agricultural regions in Africa.
Distinguished from most hyperspectral Anomaly Detection(AD) methods based on trainable parameter networks, the recently proposed method called AETNet eliminates the need for parameter adjustments or retraining on new test scenes by training an anomaly enhancement network on background data with false anomalies. In this letter, we achieve this by proposing a novel training and inference framework that enhances the network's background spectral feature extraction capability without any data augmentation. During training on background data, the complete network is trained using the reverse distillation framework with a spectral feature alignment mechanism to improve the network's background feature expressiveness. For inference, a pruned network is applied, composed solely of components most relevant to expressing features in the spectral dimension. This effectively reduces redundant information, enhancing both inference efficiency and anomaly detection accuracy. Experimental results demonstrate that our method outperforms state-of-theart methods on the HAD100 dataset, striking an optimal balance between detection accuracy and inference speed. Our code is available at https://github.com/cristianoKaKa/FERD.
California's escalating water shortage, aggravated by ongoing climate change and persistent droughts, necessitates urgent action to preserve this valuable resource. According to the United States Environmental Protection Agency, 50 percent of water used for landscape irrigation and agriculture are wasted through evaporation, wind, and runoff due to overwatering of crops1. Equally important is preventing the under watering of crops, as they can face life-threatening conditions amid California's harsh climate. To strike the delicate balance between water conservation and crop health, this paper explores a method employing soil moisture sensors for precise irrigation control. The sensors measure soil water content, enabling targeted water delivery when levels are low and immediate cessation when optimal moisture is achieved. The system is managed through an Arduino microcontroller, which efficiently regulates the irrigation process based on data gathered through the moisture sensors. The Arduino processes the information received and triggers the water supply, delivered through a pump and a hose. A sprinkler attachment at the end of the hose ensures even water distribution across all plant areas, effectively preventing overwatering in any specific spot. The results indicate over a 45 percent decrease in water use while demonstrating healthier plants. This approach presents a promising solution to California's water scarcity while ensuring sustainable crop growth and efficient resource consumption. The future plans involve using solar energy to power the device's batteries and incorporating artificial intelligence (AI) technologies to detect various factors such as plant species, soil type, terrain, and real-time weather conditions. By leveraging these advanced technologies, the research aims to transform irrigation management for enhanced water efficiency and environmental sustainability concerning California's agricultural practices.
The post-pandemic recapture of Sri Lanka's tourist industry is anticipated, and this paper suggests a hotel that employs geospatial data science for strategic placement and operational excellence. The hotel will use geospatial analytics to choose the best site, customize its hospitality offerings to meet local demands, assign resources as efficiently as possible, build strong community relationships, and navigate the competitive marketplace. The hotel will succeed in the resurgent Sri Lankan hospitality sector cheers to its data-driven strategy.
Regressive models in machine learning require regularization to balance the bias-variance tradeoff and attain realistic predictions in the real world. Two new regularization techniques, referenced as BiasWrappers, will be discussed in this paper: BiasWrapperC1 and BiasWrapperC2. BiasWrapperC1 uses a form of penalization to prevent models from consistently overshooting or undershooting. BiasWrapperC2 uses a modified layer of regression stacking to identify correlations of a regression model’s error. The techniques’ logics will be discussed through pseudocode in the context of machine learning regression. The regularization techniques are applied to machine learning models and compared with other regularization techniques through a series of carefully chosen datasets, and these metrics are used to hypothesize about the implications of these new techniques. All implementations are referenced with pseudocode in the paper, with external testing wrappers programmed in Python. An experimental study was conducted with standard regression datasets and showed the regularizations’ value propositions in multi-output data and outlier-based data.
Sea surface wave spectrum measurements are necessary for a host of basic research questions as well as for engineering and societal needs. However, most measurement techniques require great investment in infrastructure and time-intensive deployment techniques. We propose a new approach of wave measurement from standard video footage recorded by low-cost Unmanned Aerial Vehicles (UAV). We address UAV nadir imagery, which are particularly simple to obtain operationally. The method relies on the fact that optical contrast of surface gravity waves is proportional to their steepness. We present a robust methodology of regularized inversion of the optical imagery spectra, resulting in retrieval of the three-dimensional wavenumber-frequency sea surface height spectrum. The system was tested in several sea trials and in different bathymetric depths and sea state conditions. The resulting wave bulk parameters and spectral characteristics are in good agreement with collocated measurements from wave buoys and bottom-mounted acoustic sensors. Simple deployment, mobility, and flexibility in spatial coverage show a great potential of UAVs to significantly enhance the availability of wave measurements.
A multiple scattering model for passive radiative transfer (RT) in vegetation that accounts for the vertical profile of the plant structure is developed, offering advancements over the commonly-used single-layer uniform scattering models prevalent in the vegetated land surface microwave remote sensing. The proposed model takes into account the complexities of the canopy morphology with vertical heterogeneity, enabling the representation of overlapping vegetation species applicable to diverse plant types and growth stages. Additionally, it serves as a valuable tool for understanding the influence of the vegetation vertical structure on the microwave brightness temperatures. The model is constructed based on high-order solutions to the RT equations, obtained through a numerical iterative approach with an efficient interpolation scheme for algorithm acceleration. This methodology facilitates the accurate distinction of the contributions to the brightness temperature from each scattering order and scattering mechanism, ensuring a comprehensive consideration of multiple scattering effects within various vegetated scenarios. The model is validated using the SMAPVEX12 L-band forest data set, encompassing a wide range of soil moisture variations. Comparisons are made between the brightness temperatures simulated by the newly developed multiple-scattering model with a continuous profile or layered profile and those obtained from a uniform single-layer model. Results demonstrate significant improvements in the multi-layered or the continuously profiled model, showing improved agreement with the measured brightness temperatures. Furthermore, the proposed model is parameterized by matching the high-order solutions to the RT equation to the widely adopted reduced order albedo-tau formalism. The resulting equivalent parameters are linked to the geometries and the electromagnetic properties of the vegetation layer, while also incorporating the effects of multiple scattering. Comparative analysis of the equivalent parameters derived from the layered model and those derived from the single-layer model reveals that the vertical heterogeneity of the vegetation structure has a notable influence on the effective scattering albedo and it yields a value more consistent with the albedo as chosen in the SMAP/SMOS inversion algorithms. Meanwhile, the impact of the vegetation vertical profile on the effective optical thickness and the effective transmissivity of the vegetation layer is weak.These insights are essential for the retrieval of soil moisture and vegetation characteristics including the plant vertical structures in microwave remote sensing.
We address the crucial task of identifying changes in land cover using remotely sensed imagery. While most change detection methods focus on two images, we introduce an unsupervised approach that considers long image series (more than two), supporting a more nuanced differentiation between changed and unchanged areas. The proposed technique transforms input data to a new representation, capturing the target's spectral response changes over time. Areas with minimal response variation are identified as non-changing and distinguished from regions that have undergone modifications. The method further categorizes, utilizing statistical procedures, regions undergoing spatiotemporal modifications into seasonal or permanent changes. Experimental validation using simulated and real-world remote sensing image series demonstrates the effectiveness of the proposed approach.
The presence of biases has been demonstrated in a wide range of machine learning applications, yet it is not yet widespread in the case of geospatial datasets. This manuscript illustrates the importance of auditing geospatial datasets for biases with a particular focus on disaster risk management applications, as lack of local data may direct humanitarian actors to utilize global building datasets to estimate damage and the distribution of aid efforts. It is important to ensure there are no biases against the representation of vulnerable populations and that they are not missed in the distribution of aid. This manuscript audits four global building datasets (Google Open Buildings, Microsoft Bing Maps Building Footprints, Overture Maps Foundation, and OpenStreetMap) for biases with regard to Relative Wealth Index, population density, urban/rural proportions, and building size in Tanzania and the Philippines. Dataset accuracies for these two countries are lower than expected. Google Open Buildings (with a confidence above 0.7) and OpenStreetMap demonstrated the best combinations of False Negative and False Discovery, though Google Open Buildings was more consistent across tiles. The equality of opportunity was lowest for the urban/rural proportions, whereas OpenStreetMap and Overture Maps Foundation displayed particularly low equality of opportunity for population density and RWI in Tanzania. These results demonstrate that there are biases in these geospatial datasets. The types of biases are not consistent across datasets and the two study areas which emphasizes the importance of auditing these datasets for biases for new applications and study areas.Note This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.
The neutral form drag coefficient is an important parameter when estimating surface turbulent fluxes over Arctic sea ice. The form drag caused by surface features (𝑪𝒅𝒏,𝒇𝒓) dominates the total drag in the winter, but long-term pan-Arctic records of 𝑪𝒅𝒏,𝒇𝒓 are still lacking for Arctic sea ice. In this study, we first developed an improved surface feature detection algorithm and characterized the surface features (including height and spacing) over Arctic sea ice during the late winter of 2009-2019 using the full-scan laser altimeter data obtained in the Operation IceBridge mission. 𝑪𝒅𝒏,𝒇𝒓 was then estimated using an existing parameterization scheme. This was followed by applying a satellite-derived backscatter coefficient (𝝈𝒐𝒗𝒗 ) to 𝑪𝒅𝒏,𝒇𝒓 regression model to extrapolate, for the first time, 𝑪𝒅𝒏,𝒇𝒓 to the pan-Arctic scale for the entire winter season over two decades (from 1999 to 2021). We found that the surface features have a larger height and smaller spacing over multi-year ice (1.15 ± 0.21 m and 142 ± 49 m) than over first-year ice (0.90 ± 0.16 m and 241 ± 129 m). The monthly mean 𝑪𝒅𝒏,𝒇𝒓 increases through the winter, from 0.2 × 10 −3 in November to 0.4-0.5 × 10 −3 in April. The central Arctic has the largest 𝑪𝒅𝒏,𝒇𝒓 (up to 2 × 10 −3), but experienced a drop of ~50% in the period from 2001/2002 to 2008/2009. The interannual fluctuations in 𝑪𝒅𝒏,𝒇𝒓 are strongly linked to the variability of sea ice thickness and deformation, and the latter has become increasingly important for 𝑪𝒅𝒏,𝒇𝒓 since 2009.
Earth Explorer 10 mission Harmony will consist of two satellites that fly in formation with Sentinel-1. It will operate as a multistatic radar in which Sentinel-1 transmits signals and all three satellites receive signals from different lines-of-sight. To prepare for Harmony and other possible future bistatic missions, transforms are derived to map the ocean-wave spectrum into bistatic synthetic aperture radar (SAR) spectra. The SAR mapping follows the standard derivation using the multi-dimensional characteristic function, but with adjustments for the modulation transfer functions compared to the monostatic case. This paper focuses on the SAR modulations caused by velocity bunching as it is the dominant distortion mechanism. We argue that a multistatic system, such as Harmony, leads to an inversion that constrains the real aperture radar (RAR) response on a scene-by-scene basis. A benefit of having additional receivers for wave-spectra estimation is that the three lines-of-sight enables to capture a larger fraction of the wave spectrum. Improvements are especially expected in high wind speed conditions such as tropical cyclones, where large energetic surface motions strongly deteriorate the (azimuth) resolution of the SAR data. Enhanced directional wave spectral characteristics will further help to improve the interpretation of the new bistatic Harmony high-resolution scatter and Doppler combined directional measurements .
Until recently, intensity modulations in synthetic aperture radar (SAR) altimetry waveform tails have been considered a nuisance for geophysical-parameter retrieval. These modulations are actually predictable and might be exploited using a spectral analysis of the waveform tails. After , a more elaborated analysis is performed to improve the interpretation of these SAR altimeter spectra. A fast numerical model is developed to explain the modulation mechanisms in focused SAR altimetry waveform tails. Using numerical solutions, standard analytical closed-form solutions, are demonstrated to be invalid to retrieve ocean-wave-spectra retrievals from nadir altimeters. Although not valid, a closed-form derivation provides intuitive insights about the information contained in a SAR altimetry cross-spectrum. Under moderate environmental conditions (significant wave heights of ∼2 m), a closed-form solution might still be useful to infer swell-wave spectra from swath-altimetry SAR spectra, like those of the proposed Sentinel-3 Next Generation Topography mission. Comparable to side-looking SAR ocean processing, the cross-spectral analysis for nadir signals reduces noise and might remove the 180-degree ambiguity of the wave direction. Since the synthetic aperture length of nadir altimeters is larger than sidelooking SARs, sublook processing can be performed to compute multiple cross-spectra for the same scene. With a slightly changing observation geometry, resulting cross-spectra reveal slightly different parts of the ocean-wave spectrum. The resulting cross-spectral stack can thus be used to improve the retrieval of ocean-wave parameters. Retrieved ocean-wave parameters shall then enhance the sampling of the global wave field, but also serve to advance more consistent sea-state-bias corrections.
Multi-label aerial image classification is a fundamental yet complex task in remote sensing interpretation that aims to identify multiple labels in a single image. In this letter, we propose a Label-Guided Cross-Modal Attention (L-GCMA) network, which first introduces a novel approach to enriching the semantic information of labels and utilizes the multi-head attention module to extract diverse features. The proposed method consists of two components before the cross-modal attention. Firstly, the visual features of the image are obtained using a transformer encoder. Additionally, to capture the rich semantic relationship of the scene, we design a Label-Sentence Mapping Attention (L-SMA) module. This module performs word embedding encoding on the labels and applies BERT encoding on the sentences, followed by multi-head attention to extract comprehensive inter-and intra-class relationships for the labels, specifically obtaining label-scene text features. Subsequently, by treating the text features as a query, the visual features and text features are combined using cross-modal attention. This progressive integration narrows the semantic gap between vision and text, facilitating accurate label recognition. Our proposed L-GCMA consistently achieves state-of-the-art performance on the multi-label aerial image classification task, as demonstrated by extensive experiments on two visual benchmarks-the UCM multi-label dataset and the AID multi-label dataset.
Crater detection is one of the most important methods for planetary exploration. However, complex backgrounds can confuse crater detection, and a large number of small craters will lose features during the training process. To address these problems, we propose a new DEtection TRansformer (DETR) variant network for crater detection called Crater-DETR. First, we design the Correspond Regional Attention Upsample (CRAU) and Pooling (CRAP) operators by computing cross-attention between local features at different scales, which tackle the problem of foreground-background confusion caused by the loss of features after multiple downsampling for small craters. Then, some two-stage DETR variants have the issue of weak supervision in the Transformer Encoder. To alleviate this problem, we propose the Dense Auxiliary Head Supervise (DAHS) training, which could enhance the feature learning ability of the Encoder. Next, Automatic DeNoising (ADN) training is proposed to solve the problem of sparse positive queries in the Decoder to improve the decoding capability. Finally, we present a Small Object Stable IoU (SOSIoU) Loss to optimize the training process since the matching process is more unstable in small craters compared to other sizes of craters. The extensive experiments based on the DACD and the AI-TOD datasets show that Crater-DETR achieves state-of-the-art performance, especially in small craters detection.
State-of-the-art object detection methods applied to satellite and drone imagery largely fail to identify small and dense objects. One reason is the high variability of content in the overhead imagery due to the terrestrial region captured and the high variability of acquisition conditions. Another reason is that the number and size of objects in aerial imagery are very different than in the consumer data. In this work, we propose a small object detection pipeline that improves the feature extraction process by spatial pyramid pooling, cross-stage partial networks, heatmap-based region proposal network, and object localization and identification through a novel image difficulty score that adapts the overall focal loss measure based on the image difficulty. Next, we propose novel contrastive learning with progressive domain adaptation to produce domain-invariant features across aerial datasets using local and global components. We show we can alleviate the degradation of object identification in previously unseen datasets. We create a first-ever domain adaptation benchmark using contrastive learning for the object detection task in highly imbalanced satellite datasets with significant domain gaps and dominant small objects from existing satellite benchmarksâ\euro”the proposed method results in up to a 7.4% increase in mAP performance measure over the best state-of-art.Â
Hyperspectral Imaging (HSI) provides detailed spectral information for each pixel in an image, which involves acquiring images at numerous narrow and contiguous wavelength bands. Comprehensive spatial and spectral information deposits in hyperspectral images acquired by sensors, cameras, and various data acquisition sources lead to a wide range of applications across multiple fields from agriculture, and environment to biology. Various Image Processing and Artificial Intelligence algorithms have been developed periodically to analyze the data acquired through HSI. This review paper presents a comprehensive analysis of HSI focusing on its various aspects and potential implications. We explore detailed applications and key algorithms of HSI and discuss the associated advancements and challenges. Through an extensive literature review, we identify the state of research and methodologies related to HSI. Our study covers a wide range of HSI applications such as Earth Sciences, Exploration, Monitoring, Agriculture, Security, Conservation, Security, Healthcare, and Medical Imaging, and how Hyperspectral Imaging algorithms benefit these applications. Additionally, we discuss emerging trends and future directions in HSI providing insights into the promising avenues for further research.
In recent years, the likelihood of wildfire occurrence has increased in many North American communities as changes in climate have led to longer, more deadly fire seasons. Â Many Americans, especially those living in Western states, have reported frequent drought and wildfire conditions, leading to an increased need for a modeling program to assess wildfire risk at a low computational cost. The research objective of this paper was to develop a machine learning model capable of producing real-time wildfire risk assessments using five geospatial datasets: Land Fire Mean Return, Annual Precipitation, Sentinel-2 Imagery, Land Cover, and Moisture Deficit & Surplus. To create the model, three separate machine learning architectures were implemented (U-Net, DeepLabV3, and the Pyramid Scene Parsing Network) and then applied to the study area of San Bernardino County, CA for the year 2020. In addition, this study demonstrated a proof of concept for further inquiry into combining artificial intelligence and geospatial datasets to create useful insights. Â