Application Of Satellite Imagery Research Articles

Application of satellite imagery for vector-borne disease monitoring in sub-Saharan Africa: An overview

This work presents an overview of the application of satellite imagery or remote sensing (RS) data for vector-borne disease (VBD) monitoring in sub-Saharan Africa. We discussed the different vector-borne disease conditions that are prevalent and endemic in parts of sub-Saharan Africa and how satellite imagery (RS data) can be used in monitoring these conditions. Key disease conditions considered include malaria, human African trypanosomiasis, leishmaniasis, lymphatic filariasis, loa loa filariasis, rift valley fever, dengue, yellow fever, and rickettsioses. Furthermore, we explored some of the current ways remote sensing data and geographical information systems (GIS) are being applied to monitoring these diseases. We discuss the efficacy of using strong spatial modelling techniques combined with RS data to enhance our comprehension of the role that environmental conditions play in influencing VBD vectors and transmission alongside the utilization of GIS/RS technologies for vector-borne illness surveillance, prevention, and control. Finally, we discussed how drone technology and new remote sensing platforms can provide improved monitoring and the impact of such improvements from a population health standpoint.

Decoupled Training: Return of Frustratingly Easy Multi-Domain Learning

Multi-domain learning (MDL) aims to train a model with minimal average risk across multiple overlapping but non-identical domains. To tackle the challenges of dataset bias and domain domination, numerous MDL approaches have been proposed from the perspectives of seeking commonalities by aligning distributions to reduce domain gap or reserving differences by implementing domain-specific towers, gates, and even experts. MDL models are becoming more and more complex with sophisticated network architectures or loss functions, introducing extra parameters and enlarging computation costs. In this paper, we propose a frustratingly easy and hyperparameter-free multi-domain learning method named Decoupled Training (D-Train). D-Train is a tri-phase general-to-specific training strategy that first pre-trains on all domains to warm up a root model, then post-trains on each domain by splitting into multi-heads, and finally fine-tunes the heads by fixing the backbone, enabling decouple training to achieve domain independence. Despite its extraordinary simplicity and efficiency, D-Train performs remarkably well in extensive evaluations of various datasets from standard benchmarks to applications of satellite imagery and recommender systems.

Attention-Based Semantic Segmentation Networks for Forest Applications

Deforestation remains one of the key concerning activities around the world due to commodity-driven extraction, agricultural land expansion, and urbanization. The effective and efficient monitoring of national forests using remote sensing technology is important for the early detection and mitigation of deforestation activities. Deep learning techniques have been vastly researched and applied to various remote sensing tasks, whereby fully convolutional neural networks have been commonly studied with various input band combinations for satellite imagery applications, but very little research has focused on deep networks with high-resolution representations, such as HRNet. In this study, an optimal semantic segmentation architecture based on high-resolution feature maps and an attention mechanism is proposed to label each pixel of the satellite imagery input for forest identification. The selected study areas are located in Malaysian rainforests, sampled from 2016, 2018, and 2020, downloaded using Google Earth Pro. Only a two-class problem is considered for this study, which is to classify each pixel either as forest or non-forest. HRNet is chosen as the baseline architecture, in which the hyperparameters are optimized before being embedded with an attention mechanism to help the model to focus on more critical features that are related to the forest. Several variants of the proposed methods are validated on 6120 sliced images, whereby the best performance reaches 85.58% for the mean intersection over union and 92.24% for accuracy. The benchmarking analysis also reveals that the attention-embedded high-resolution architecture outperforms U-Net, SegNet, and FC-DenseNet for both performance metrics. A qualitative analysis between the baseline and attention-based models also shows that fewer false classifications and cleaner prediction outputs can be observed in identifying the forest areas.

Insights for Sea Outfall Turbid Plume Monitoring with High-Spatial-Resolution Satellite Imagery Application in Portugal

Coastal municipalities and industries often discharge poorly treated wastewater into proximate marine and estuarine environments. The urban and/or effluent input can lead to eutrophication and lower water quality, as it holds high concentrations of nutrients and pollutants. One widely applied tool to increase effluent dispersion and direct it away from coastal areas, thus causing fewer impacts on human activities, is sea outfall. In Aveiro, Portugal, the São Jacinto sea outfall construction was completed in 1998; however, limited literature regarding the sea outfall’s monitoring using satellite data is available. The methodology in this study involved collecting four years’ worth (2016–2019) of satellite data to investigate visible traces of the interaction between the S. Jacinto sewage water mass and the Ria de Aveiro lagoon ecosystem using ocean color and spectral analysis, and producing new qualitative data regarding sea outfall plume dispersion monitoring through high-resolution Sentinel-2 imagery. The results showed a clear spectral signature of the sewage water mass, and a seasonal pattern was observed in which the plume was more evident in winter and autumn. Additionally, the coastal longshore current and the Aveiro lagoon’s runoff were able to restrict the marine outfall’s dispersion superficially. Ocean color data were revealed to be a factual and cost-effective tool to monitor the plume water. Finally, an exchange between the marine outfall water mass and Ria de Aveiro lagoon could happen in high tide under northern wind conditions. Therefore, it is important to monitor the water quality to ensure the coastal ecosystem’s good environmental state.

EXPERIENCE OF USING PLANET EARTH OBSERVATION DATA IN RETRAINING COURSES FOR EDUCATORS IN THE JUNIOR ACADEMY OF SCIENCES OF UKRAINE

This paper discusses the background and presents the results of the first in Ukraine advanced training course for educators in Remote Sensing, “Fundamentals of Remote Sensing: History and Practice”, organized by the Geographic Information Systems and Remote Sensing Laboratory of the National Center, “Junior Academy of Sciences of Ukraine”. The Junior Academy of Sciences of Ukraine is a system of extracurricular education committed to the development and implementation of science education techniques. Currently, the Academy has more than 250,000 students working in 64 scientific areas. In 2018, the Junior Academy of Sciences of Ukraine received the status of Category 2 Science Education Center under the auspices of UNESCO and joined the network of Copernicus Academies. The Junior Academy of Sciences of Ukraine has been promoting the application of satellite imagery in education for over ten years. One of the outputs of this activity is the advanced training course for educators. This paper describes the online training course that was held in spring 2021 and 2022. The course was open to all interested teachers. A total of 41 educators from 13 regions of Ukraine have completed the training course in the span of two years. The paper describes the stages of special course organization, topics, and tools used in the lectures and case studies, including an overview of the data sources, such as EO-Browser, ArcGIS Online, Google my Maps, NASA Worldview and Google Earth, etc. The paper presents a detailed description of one of the hands-on case studies of water bodies in EO-Browser. It was established that the conducted special course promoted the educators’ interest in bringing Earth observation to classrooms. Increased number of GIS and Remote Sensing after-school clubs in the territorial branches of the Junior Academy of Sciences (from 1 in Kyiv as of 2019 to 9 in the regions as of 2022, namely in Poltava, Sumy, Volyn, Kyiv, Kirovohrad, Zakarpattia, Lviv, Kharkiv and Chernivtsi) is indicative of the course’s effectiveness.

Review on Various Image Processing Techniques in Satellite Imagery Applications

Review on Various Image Processing Techniques in Satellite Imagery Applications

A review of machine learning and satellite imagery for poverty prediction: Implications for development research and applications

AbstractThe field of artificial intelligence is seeing the increased application of satellite imagery to analyse poverty in its various manifestations. This nascent but rapidly growing intersection of scholarship holds the potential to help us better understand poverty by leveraging big data and recent advances in machine vision. In this study, we statistically analyse the literature in the expanding field of welfare and poverty predictions from the combination of machine learning and satellite imagery. Here, we apply an integrative review method to extract key data on factors related to the predictive power of welfare. We found that the most important factors correlated to the predictive power of welfare are the number of pre‐processing steps employed, the number of datasets used, the type of welfare indicator targeted and the choice of AI model. Studies that used stock measure indicators (assets) as targets achieved better performance—17 percentage points higher—in predicting welfare than those that targeted flow measures (income and consumption) ones. Additionally, we found that the combination of machine learning and deep learning significantly increases predictive power—by as much as 15 percentage points—compared to using either alone. Surprisingly, we found that the spatial resolution of the satellite imagery used is important but not critical to the performance as the relationship is positive but not statistically significant. These findings have important implications for future research in this domain and for anyone aspiring to use the methodology.

An assessment of high temporal frequency satellite data for historic environment applications. A case study from Scotland

AbstractThis paper assesses the value of high temporal frequency satellite data with various spatial sampling resolutions for multi‐scalar historic environment survey and management use cases in Scotland, specifically for broad‐brush landscape characterisation, for monitoring the condition of monuments and for the discovery of otherwise unknown sites. Dealing with a part of the world where applications of satellite imagery are almost entirely unexplored, this study takes a real‐world approach, which foregrounds the purpose at hand rather than presenting a case study from an optimal setting. The study highlights the importance of detailed imagery to support interpretation in some instances, and the challenges of obtaining time‐critical optical imagery in a part of the world that experiences significant periods of cloud cover. The real‐world availability of data in such settings is assessed, highlighting that even with daily revisits, useable imagery cannot be guaranteed. The implications of current and past tasking patterns for availability of high‐resolution data now and in the future are discussed. The study identifies the complementary roles that satellite imagery can fulfil, while identifying the limitations that remain to fuller applications of such data, in a study that will be relevant to many parts of Europe and beyond.

How the application of satellite imagery offers new opportunities to explore China’s urbanisation and development

How the application of satellite imagery offers new opportunities to explore China’s urbanisation and development

Spatial association between COVID-19 incidence rate and nighttime light index.

This study statistically identified the localised association between socioeconomic conditions and the coronavirus disease 2019 (COVID-19) incidence rate in Thailand on the basis of the 1,727,336 confirmed cases reported nationwide during the first major wave of the pandemic (March-May 2020) and the second one (July 2021-September 2021). The nighttime light (NTL) index, formulated using satellite imagery, was used as a provincial proxy of monthly socioeconomic conditions. Local indicators of spatial association statistics were applied to identify the localised bivariate association between COVID-19 incidence rate and the year-on-year change of NTL index. A statistically significant negative association was observed between the COVID-19 incidence rate and the NTL index in some central and southern provinces in both major pandemic waves. Regression analyses were also conducted using the spatial lag model (SLM) and the spatial error model (SEM). The obtained slope coefficient, for both major waves of the pandemic, revealed a statistically significant negative association between the year-on-year change of NTL index and COVID-19 incidence rate (SLM: coefficient= âˆ'0.0078 and âˆ'0.0064 with P<0.001 and 0.056, respectively; and SEM: coefficient= âˆ'0.0086 and âˆ'0.0083 with P=0.067 and 0.056, respectively). All of the obtained results confirmed the negative association between the COVID-19 pandemic and socioeconomic activity revealing the future extensive applications of satellite imagery as an alternative data source for the timely monitoring of the multidimensional impacts of the pandemic.

Mapping Large-Scale Plateau Forest in Sanjiangyuan Using High-Resolution Satellite Imagery and Few-Shot Learning

Monitoring the extent of plateau forests has drawn much attention from governments given the fact that the plateau forests play a key role in global carbon circulation. Despite the recent advances in the remote-sensing applications of satellite imagery over large regions, accurate mapping of plateau forest remains challenging due to limited ground truth information and high uncertainties in their spatial distribution. In this paper, we aim to generate a better segmentation map for plateau forests using high-resolution satellite imagery with limited ground-truth data. We present the first 2 m spatial resolution large-scale plateau forest dataset of Sanjiangyuan National Nature Reserve, including 38,708 plateau forest imagery samples and 1187 handmade accurate plateau forest ground truth masks. We then propose an few-shot learning method for mapping plateau forests. The proposed method is conducted in two stages, including unsupervised feature extraction by leveraging domain knowledge, and model fine-tuning using limited ground truth data. The proposed few-shot learning method reached an F1-score of 84.23%, and outperformed the state-of-the-art object segmentation methods. The result proves the proposed few-shot learning model could help large-scale plateau forest monitoring. The dataset proposed in this paper will soon be available online for the public.

Application of satellite imagery to update depth-area-volume relationships in reservoirs in the semiarid region of Northeast Brazil

ABSTRACT Reservoirs are the primary source of water supply in the semiarid region of Pernambuco state, Brazil, because of the constant water scarcity affecting this region. Knowledge of the amount of water available is essential for the effective management of water resources. The volume of water stored in the reservoirs is calculated using the depth-area-volume relationship. However, in most reservoirs in the semiarid region, this relationship is currently out of date. Therefore, the objective of this study was to explore the potential and limitations of the application of the ISODATA unsupervised classification method to calculate the depth-area-volume relationships of reservoirs in the semiarid region of Pernambuco, Brazil. The application of the ISODATA method was evaluated in three reservoirs in the state of Pernambuco, i.e., Poço da Cruz, Barra do Juá, and Jucazinho. The results were compared with the updated curves of reservoirs obtained from bathymetry and recent LiDAR surveys. The ISODATA method presented satisfactory results for the three reservoirs analyzed. The mean absolute error of the volume in Poço da Cruz and Barra do Juá was lower than 1% of the maximum capacity. The use of the ISODATA method meant that the surface area underestimation tendency in the Poço da Cruz reservoir was less than when spectral indices were used.

How Much Can We See from a UAV-Mounted Regular Camera? Remote Sensing-Based Estimation of Forest Attributes in South American Native Forests

Data collection from large areas of native forests poses a challenge. The present study aims at assessing the use of UAV for forest inventory on native forests in Southern Chile, and seeks to retrieve both stand and tree level attributes from forest canopy data. Data were collected from 14 plots (45 × 45 m) established at four locations representing unmanaged Chilean temperate forests: seven plots on secondary forests and seven plots on old-growth forests, including a total of 17 different native species. The imagery was captured using a fixed-wing airframe equipped with a regular RGB camera. We used the structure from motion and digital aerial photogrammetry techniques for data processing and combined machine learning methods based on boosted regression trees and mixed models. In total, 2136 trees were measured on the ground, from which 858 trees were visualized from the UAV imagery of the canopy, ranging from 26% to 88% of the measured trees in the field (mean = 45.7%, SD = 17.3), which represented between 70.6% and 96% of the total basal area of the plots (mean = 80.28%, SD = 7.7). Individual-tree diameter models based on remote sensing data were constructed with R2 = 0.85 and R2 = 0.66 based on BRT and mixed models, respectively. We found a strong relationship between canopy and ground data; however, we suggest that the best alternative was combining the use of both field-based and remotely sensed methods to achieve high accuracy estimations, particularly in complex structure forests (e.g., old-growth forests). Field inventories and UAV surveys provide accurate information at local scales and allow validation of large-scale applications of satellite imagery. Finally, in the future, increasing the accuracy of aerial surveys and monitoring is necessary to advance the development of local and regional allometric crown and DBH equations at the species level.

Tracing suspended matter in Tiab estuary applying ANN and Remote sensing

Knowledge about deposition and sedimentation in estuaries is important since these environments are favorable domain for human activity. Variety of hydrodynamic parameters causes sedimentation/deposition in an estuary among them tidal current. It is therefore necessary to monitor these processes in estuaries and provide strategies to control the rate of sedimentation/deposition. Several methods are available to measure total suspended solids (TSS). Using combined methods of direct and indirect suspended sediment measurements in this research, we achieved a reliable method to study the suspended sediment status in an estuary located northern of the Persian Gulf, known as Tiab estuary. Field measurements were performed as two scheduled cruises one in October and one in November 2015. Current velocity and turbidity were recorded using RCM9 device and optical backscatter sensor. Water samples were collected using Niskin Bottle. Measurements were carried out both along the estuary and stationery. Artificial Neural Network (ANN) in combination with Linearity method has been applied to reproduce sufficient SS data. The ANN feed-forward back-propagation provides reliable TSS values with the RMSE and correlation coefficient of 0.137 and 0.943, respectively. Results show that the estuary is flood dominated, thus in general the estuary is under constant sedimentation. Applicability of satellite imagery of Sentinel to trace TSS in this estuary is investigated and improved. The difference between the SST rate in the four tidal phases, including high and low tide and flood and ebb phase, was examined using the NSMI-index. We found the average NSMI index in high tide is 0.216, in the low tide is 0.278, in flood phase is 0.302, and in ebb phase is 0.295. Both field and satellite data show that the concentration of TSS in the ebb phase increases, and in the low and high tide decreases. Both results also revealed that the accumulation of TSS in the meandered area is relatively high.

Remote sensing-based operational modeling of fuel ignitability in Hyrcanian mixed forest, Iran

To date, the efficiency and effectiveness of early warning systems of satellite imagery for preventing and mitigating wildfire remain a challenging issue. The heat of pre-ignition ( $$Q_{{{\text{ig}}}}$$ ) can be an index of fire likelihood, which is further enhanced with remotely sensed data, active fire data, and fuels information for operational application of satellite imagery in fire early warning systems. $$Q_{{{\text{ig}}}}$$ is a prerequisite for forest fires by the side of ignition sources and weather. This study analyzed the effect of $$Q_{{{\text{ig}}}}$$ variation on fire occurrences to develop a remote sensing-based initial fire likelihood index for identifying areas that have a high probability of fire. In this study, $$Q_{{{\text{ig}}}}$$ of Rothermel’s fire spread model daily data is retrieved at 1 km pixels from MODIS data. MODIS active fire products were used to interpret the $$Q_{{{\text{ig}}}}$$ of fuels for 10 days before the days of fire occurrences in November 2010 to determine the pre-fire conditions. A formula for converting $$Q_{{{\text{ig}}}}$$ into an initial fire likelihood index (IFLI) was then used by binary logistic regression method. Analyses show that there was a positive association between suggested IFLI and fire occurrences during the study period with a fair diagnostic accuracy of 92%, and 80% for dead and live fuels, respectively. Mann–kendall test suggested that there are significant trends in the fuel moisture content time-series for both live and dead fuels. Further analysis using the Hosmer–Lemeshow test represents that the models showed an acceptable fit. The suggested IFLI is an effective tool for fire management decision-making whenever a near real-time fire likelihood is required.

THE COURSE OF FUNDAMENTALS OF REMOTE SENSING FOR EDUCATORS OF JUNIOR ACADEMY OF SCIENCES OF UKRAINE

This paper discusses the background and presents the results of the first in Ukraine advanced training course for educators in Remote Sensing, “Fundamentals of Remote Sensing: History and Practice”, organized by the Geographic Information Systems and Remote Sensing Laboratory of the National Center, “Junior Academy of Sciences of Ukraine”. The Junior Academy of Sciences of Ukraine is a system of extracurricular education committed to the development and implementation of science education techniques. Currently, the Academy has more than 250,000 students working in 64 scientific areas. In 2018, the Junior Academy of Sciences of Ukraine received the status of Category 2 Science Education Center under the auspices of UNESCO and joined the network of Copernicus Academies. The Junior Academy of Sciences of Ukraine has been promoting the application of satellite imagery in education for over ten years. One of the outputs of this activity is the advanced training course for educators. This paper describes the online training course that was held in spring 2021. The course was open to all interested teachers. 21 educators from 11 Oblasts of Ukraine have completed the training course. The majority of the course participants were teachers of Geography. Environmental science, biology, chemistry, physics, and computer science experts have also participated in the training course. In addition, the course participants included four associate professors and two professors from higher educational institutions, and one PhD in Geography. From a gender perspective, 66.6 % of the registered course participants were female and 33.3 % — male. The effectiveness of the course was evaluated based on the participants’ responses collected in two surveys and a concluding interview. The paper describes the stages of special course organization, topics, and tools used in the lectures and case studies, including an overview of the data sources, such as EO Browser, ArcGIS Online, and Google Earth, etc. The paper presents a detailed description of one of the hands-on case studies of water bodies in EO Browser.

Automatic Detection of Dominant Crop Types in Poland Based on Satellite Images

Abstract The assumption of the European Union Common Agricultural Policy is to maintain good agricultural practices for sustainability in the environment. A number of requirements are imposed on farmers, including the maintenance of permanent grassland, fallow land or crop diversification. To meet these requirements, the European Union guarantees subsidies, but at the same time fields must be monitored focusing on crop identification. The limitation of field inspection and substituting it with crop recognition using satellite images could increase the effectiveness of this procedure. The application of satellite imagery in automatic detection and identification of dominant crops over a large area seems to be technically and economically sound. The paper discusses the concept and the results of automatic classification based on a Random Forests classifier performed on multitemporal images of Sentinel-2 and Landsat-8. A test site was established in a complex agricultural structure with long and narrow parcels in the south-eastern part of Poland. Time-series images acquired during the growing season 2016 were used for multispectral classification in different configurations: for Sentinel-2 and Landsat-8 separately and for both sensors integrated. Different Random Forests approaches and post-processing methods were examined based on independent data from farmers’ declarations records, reaching the best accuracy of over 90% for crops like winter or spring cereals. Overall accuracy of the classification ranged from 72% to 91% depending on the classification variant. The elaborated scheme is novel in the context of Polish complex agricultural structure and smallholders.

Application of Satellite Imagery and Water Indices to the Hydrography of the Cetina RiverBasin (Middle Adriatic)

The paper gives a brief description of the remote sensing method used for the identification and extraction of water surfaces. Landsat 8 and Sentinel 2 satellite imagery was used to separate land from bodies of water in the complex karst area surrounding the Croatian Cetina River, flowing into the Adriatic Sea. Water indexing methods are presented in detail. The most frequently used water indices were selected: NDWI, MNDWI, AWEI_nsh, AWEI_sh, WRI and LSWI, and their results compared. The combination of satellite imagery and calculated water indices is concluded to be very useful for the identification and mapping of the area and banks of lakes, riverine zones, river mouths and the coastline in the coastal zone. Landsat 8 satellite imagery is slightly inferior to Sentinel 2 due to lower image resolution. The best results were obtained with the NDWI water index and the worst with LSWI.

Feature points-based image registration between satellite imagery and aerial images of agricultural land

Rapid advancement in remote sensing sensors has resulted in an enormous increase in the use of satellite imagery (SI) and images taken from unmanned aerial vehicles (UAVs) in a wide range of remote sensing applications. These applications include urban planning, environment monitoring, map updating, change detection, and precision agriculture. This paper focuses on an agricultural application of SI and UAV images. SI-UAV images possess high temporal, textural, and intensity differences due to rapid changes in agricultural crops with the passage of time. Feature points such as scale invariant feature transform (SIFT), oriented FAST and rotated BRIEF (ORB), and speeded-up robust features (SURF) are not invariant to such differences and underperform in SI-UAV image registration. To deal with this problem, we propose a new method that combines the strength of nearest neighbor (NN) and brute force (BF) descriptor matching strategies to register SI?UAV images. The proposed method is named NN-BF. For NN-BF first corresponding feature point descriptor matches are identified between SI-UAV images of the training set with overlap error. Then the corresponding descriptors are matched with the descriptors of SI images of the test set with NN strategy. The resulting descriptor matches are then further matched with the descriptors of UAV images of the test set using BF strategy. Finally, the descriptor matches obtained are processed with RANSAC to remove outliers and estimate a homography for image registration. Experiments are performed on an agricultural land image dataset. The experimental results show that the NN-BF method improves SIFT, SURF, and ORB feature point performances and also outperforms recently proposed feature matching strategies for remote sensing images. SIFT on average obtains 6.1% and 18.9% better precision scores than SURF and ORB with NN-BF, respectively. SIFT also obtains lower root mean square error than SURF and ORB with NN-BF.

Satellite-Imagery Geometric Accuracy Improvement Based on Direct Correction of Dominant Coefficients

An important and difficult point in the application of satellite imagery is refining the positioning model and improving the geometric accuracy. In this study, we focus on improvement in geometric accuracy and develop a new rational function model (&lt;small&gt;RFM&lt;/small&gt;) refinement method. First, we derive the conversion relationship between the rigorous sensor model and the &lt;small&gt;RFM&lt;/small&gt;, based on which we illustrate the approximate meaning of the zero-order and first-order terms of the rational polynomial coefficients (&lt;small&gt;RPCs&lt;/small&gt;). Second, the correlation problem between &lt;small&gt;RPCs&lt;/small&gt; and the influence of individual &lt;small&gt;RPCs&lt;/small&gt; on geometric positioning accuracy are analyzed and verified. The dominant coefficients that determine geolocation are then identified. Finally, a new &lt;small&gt;RFM&lt;/small&gt; refinement method based on direct correction of the dominant coefficients is proposed and validated. The experiments, conducted with &lt;small&gt;ZY3-02&lt;/small&gt; satellite imagery, indicate that the proposed method can effectively improve the geometric accuracy of satellite images.