In this presentation, we will report our recent efforts in achieving high performance in Antimonides Type-II Superlattice (T2SL) based infrared photodetectors using the Barrier Infrared Detector (BIRD) architecture. The High Operating Temperature (HOT) BIRD Focal Plane Arrays (FPAs) offer the same high performance, uniformity, operability, manufacturability, and affordability advantages as InSb. However, Mid-Wavelength Infrared (MWIR) HOT-BIRD FPAs can operate at significantly higher temperatures (⪆150K) than InSb FPAs (typically 80K). Moreover, while InSb has a fixed cutoff wavelength (~5.4 μm), the HOT-BIRD offers a continuous adjustable cutoff wavelength, ranging from ~4 μm to ⪆15 μm, and is therefore also suitable for Long Wavelength Infrared (LWIR) as well. The LWIR detectors based on the BIRD architecture has also demonstrated significant operating temperature advantages over those based on traditional p-n junction designs. HyTI (Hyperspectral Thermal Imager) and c-FIRST (compact Fire Infrared Radiance Spectral Tracker) based on JPL’s T2SL BIRD FPAs. Based on III-V compound semiconductors, the BIRD FPAs offer a breakthrough solution for the realization of low cost (high yield), high-performance FPAs with excellent uniformity and pixel-to-pixel operability.

The aim of PAN-sharpening is to fuse a Multispectral (MS) image and a Panchromatic (PAN) image into a High-Resolution Multispectral (HRMS) image. The spatial resolution of the HRMS image is extracted from the PAN image, while the spectral resolution is extracted from the MS image. In this paper, a PAN-sharpening model based on the gradient constraint and Laplacian regularization is proposed. The objective function, which is a convex optimization problem, aims to minimize three least-square terms: (1) spectral constraint, (2) spatial constraint, and (3) image regularization. In experiments, the proposed method not only demonstrates better visual quality but also shows improvement in many quality metric evaluations.

Normalized Difference Vegetation Index (NDVI) is one of the common vegetation indices used to assess forest health. This includes the use of several wavelengths in the electromagnetic spectrum. Time Series analysis was the common process used in studying the temporal behavior of this remote sensing data. This process provides straightforward information about how NDVI changes over time and only provides minimal information. In this study, Recurrence Quantification Analysis (RQA) provides another approach to studying NDVI. Recurrence quantification analysis quantifies the number of recurrences of a state or data point in the time series using a Recurrence plot as a graphical representation. Based on the patterns of the recurrence plots, the NDVI on sites one to five are all deterministic periodic. The datasets manifest long diagonal lines, a distinctive feature of a periodic behavior. This recurrence quantification analysis method can assess other biogeophysical parameters that exhibit periodic behavior that depends on different weather parameters.

The Cross-track Infrared Sounder (CrIS) on the Joint Polar Satellite System (JPSS) satellites is a Fourier transform spectrometer, providing sounding (temperature and humidity) and trace gas the of the atmosphere with 2221 spectral channels along three infrared bands. This paper presents a Machine Learning (ML) method to retrieve the methane in the middle troposphere from CrIS spectra. Different from traditional physical retrieval methods, the main idea of the ML-based approach is to use a Neural Network (NN) to approximate the complex inverse function that maps the target trace gas concentration to the CrIS radiances measured based on the training datasets. The authors utilize the Thermodynamic Initial Guess Retrieval (TIGR) dataset coupling with a three-dimensional chemical-transport model outputs (Global Greenhouse Gas Reanalysis) to build the training datasets through radiative transfer model calculations. The training dataset covers a large range not only of the concentration of the target trace gas but also of the auxiliary parameters on the state of the atmosphere. A deep residual neural network (ResNet) is trained to determine model parameters (weights for each node). The preliminary results are encouraging and indicate that the AI-based method has the ability to retrieve the tropospheric methane from CrIS data. In proposed future work, we will use the real CrIS spectral as inputs to estimate the tropospheric methene. These newly developed products will be compared with the existing sounding products from physical methods as well as those directly measured from ground and aircraft.

This paper presents the initial results obtained from airborne data cubes collected by RedEye-1, a cost-effective Short-Wave Infrared (SWIR) imager designed for measuring methane (CH₄) and carbon dioxide (CO₂) concentrations in the atmosphere. RedEye-1 operates within the spectral range of 1588-1673 nm with an approximate spectral resolution of 0.35 nm. The imager completed its first airborne mission, capturing data cubes over coal mines in New South Wales, Australia. This paper presents and interprets the data cubes acquired while detailing some challenges encountered during the mission. The inversion model estimated the CH₄ concentration near 1645 nm to be approximately 5229 ppb, compared to the standard atmospheric CH₄ level, implying significant CH₄ emissions from the coal mine flown over.

A new method utilizing hyperspectral infrared sounders is developed to estimate Cloud Top Height (CTH) for deep convective clouds associated with hurricane, or Tropical Cyclone (TC). By analyzing measurements from the Cross-track Infrared Sounder (CrIS), and further validating with radiative transfer simulations, we found an inverted-V spectral feature in the ozone (O₃) band near 9.6 μm for thick clouds, and its depth, designated as H_index, has a strong correlation with CTH. Using a linear regression, a formula is derived to calculate the CTH based on H_index and Brightness Temperature (BT) in other three channels. This method effectively captures the cloud structure of a TC's eyewall and surrounding rainbands, with an error of -0.05 ± 0.19 km (or -0.41 ± 1.96%). Further analysis reveals that the retrieved temperature profiles near hurricane’s eye from a new single Field of View sounder products (SiFSAP) from CrIS agree reasonably well with the reanalysis data from MERRA-2 and ERA-5. This method can be easily applied to operationally monitor hurricane cloud and its development, and the estimated CTH can be also used as a-priori to improve the retrieval products.

Airborne-based active multispectral sensing becomes an essential technology in the development of smart agriculture; unfortunately, the effectiveness of airborne-based spectral sensing is affected from atmospheric turbulence and atmospheric refraction variations. Atmospheric turbulence can degrade receiving energy density and atmospheric refraction variations can induce Angle of Arrival (AOA) error and chromatic dispersion from detecting target. A multispectral sensing model through anomalous atmosphere is introduced; sensing energy dispersed by atmospheric turbulence is discussed, sensing angle of arrival error induced by atmospheric refraction is analyzed by ray tracing. To evaluate sensing performance, theoretical computing sensing results included receiving energy and AOA error under various weather conditions are reported.

Central America is one of the major regions for banana cultivation. However, due to the most of banana trees in this area are never infected by Panama disease, the major species of banana trees in Central America are still not Panama disease resistant and highly vulnerable to the threat of the disease. This disease causes yellowing, wilting, and reduced yield in banana plants, resulting in significant impacts on farmers' income and the economy of the region. In this study, Taiwan Space Agency (TASA) and the International Cooperation and Development Found (TaiwanICDF) are collaborated to develop an effective monitoring and early warning system to cope with this issue and a multi-temporal satellite image cloud storage and processing platform, as known as Taiwan Data Cube (TWDC), has been utilized. The platform majorly enables regular analysis process of satellite imagery from Formosat-5 and Sentinel-2 satellites and delivery of corresponding analyzed results to local users. These high-resolution images provide valuable surface observation data in assessing the health status of banana plants and tracking disease development trends. The project has established 15 monitoring areas in Guatemala and six in Belize, covering a total area of 200 hectares. The research demonstrates the feasibility of utilizing satellite data for large-scale plant disease monitoring and a successful collaboration with the Taiwan Technical Mission groups of TaiwanICDF in Central American. This development provides valuable information to farmers and disease control authorities, facilitating more effective disease monitoring and management practices.

Locust monitoring is crucial for preventing locust plagues. This study proposes an algorithm for detecting locust swarms across large areas with minimal labor, utilizing space-borne or airborne hyperspectral sensors. Laboratory spectral measurements were conducted to verify the proposed algorithm. A grassland environment was recreated in the laboratory and locusts were placed in it. The Locust Coverage Ratio (LCR), which is the projected area of locusts per unit area of grassland, was retrieved from the spectral reflectance measured using our algorithm. Locusts were detected by applying a threshold to the retrieved LCR, and the detection performance was evaluated using the Area Under the Curve (AUC) of the Receiver Operating Characteristics. The AUC exceeded 0.9 when locust density on the grassland was 20 locusts per square meter, corresponding to an LCR of 1% under the assumption that the projected area of each locust was 5 cm² . These results demonstrate the potential effectiveness of the proposed algorithm for large-scale locust monitoring using satellites.

Efficient disaster surveys can safeguard the compensation rights of affected farmers and serve as a critical component of market stabilization strategies. Due to climate change, Taiwan is experiencing more frequent climate disasters. The agricultural industry faces severe labor shortages making post-disaster recovery increasingly challenging. This issue may impact food supply security, disrupt prices, and threaten national security. To address this, the study applies an advanced and efficient semantic segmentation network model to the Unmanned Aerial Vehicle (UAV) captured imagery for rice lodging disaster assessment. By incorporating a rule-based multi-task learning framework, prior knowledge from physical rules constrains the classifier's learning. Preliminary results indicate that the modified model achieved a 10% above improvement in the recall rate for lodged rice compared to the original model using 2017 data, and around a 5% improvement on the transferred 2019 data. Suggesting that this study can predict rice lodging with a more interpretable model architecture and achieve better classification results.

The coastal regions of Maharashtra, India are experiencing significant shoreline changes due to both natural and anthropogenic factors. This study investigates these changes using Google Earth Engine, Digital Shoreline Analysis System (DSAS) and ArcMap, focusing on seven coastal districts of Maharashtra state namely Palghar, Thane, Mumbai Suburban, Mumbai, Raigarh, Ratnagiri, and Sindhudurg. Multi-temporal Landsat satellite images were used for extraction of historical shoreline (1993 to 2023) in cloud platform Google earth engine and further were analyzed in DSAS and ArcMap10.7. Statistical techniques like Shoreline Change Envelope (SCE) and Net Shoreline Movement (NSM) were used to measure the spatial extent of shoreline shift. The rate of change was calculated using the End Point EPR) and Linear Regression Rate (LRR) to quantify the rate of coastal erosion and accretion. Around 8416 transects were placed at 100m intervals along the coast, with approximately 32% of the transects indicating erosion, 44% accretion and remaining are in stable condition. The research revealed that Palghar experiences the most dramatic shoreline alterations among the seven coastal districts, with highest rate of erosion (EPR) -20.5 m/yr. Aside from that, Mumbai has the highest accretion rates up to a maximum (EPR) 50 m/yr. Thane, Mumbai, Ratnagiri and Sindhudurg has eroded less in comparison to Raigarh and Mumbai Suburban coast. LRR yields the average erosion at -1.13 m/yr and average accretion at 2.31 m/yr from 1993 to 2023 for 1486km long shoreline of Maharashtra. The Net areal change was estimated to be 20.43 sq.km was calculated by subtracting total area lost from total area gained over the three decades. Finally, recognizing the erosion and accretion hotspots using cloud computing, remote sensing and Geographic Information System along the larger region like Maharashtra coast can assist academicians, policymakers and stakeholders. This research also enables remote sensing and coastal researchers to understand regional shoreline dynamics.

This research aims to explore the effectiveness and functionality comparison of applying Artificial Intelligence (AI) and deep learning (DL) packages in QGIS to FORMOSAT satellite imagery. The selected AI and DL tools include Deep Neural Remote Sensing developed by PUT Vision Lab, Mapflow.ai developed by the Geoalert team, and SAGA GIS developed by the University of Göttingen. Using FORMOSAT satellite imagery as test data, these tools were evaluated for their applicability and functionality. The results show that each package has its own strengths in image processing, with specific applicability depending on the requirements of the image applications and characteristics of the imagery. This study provides practical references for the application of artificial intelligence and deep learning in remote sensing using FORMOSAT series satellite imagery.

Sugarcane and paddy rice are stable food in Thailand, since they are significant cash crops for international trade. Currently the sugarcane and paddy rice investigation carried out by Thai government is in once per two-year period. However, due to the rapid change of farmers growing behavior and the needs of market, the field survey encounters low efficacy and accuracy. This research is supported under the joint project “Paddy Field Land Use Change Detection Using Data Cube’’ between Taiwan Space Agency (TASA) and National Electronics and Computer Technology Center (NECTEC) in Thailand.

Taiwan Data Cube (TWDC) is a cloud platform used for the storage and processing of multi-sensor, multi-temporal remote sensing data. The platform is majorly an image application platform for making value-added services for user customized applications. In this study, the image analysis application for landslide detection based on AI technique in TWDC is used as an example to demonstrate how it can be shared to local user with security. Currently, to share an application properly without security consideration from an application server, the use of Application Programming Interface (API) is a standard procedure and has been widely used in information technology area. However, for the researcher who develop remote sensing applications, to build an API connectivity for local users to submit parameters and obtain results from the application in cloud environment is still a difficult job. Therefore, for generating a demonstration that can be followed for those who need an example to establish API connectivity in sharing their image processing application in TWDC, following are steps applied in this study: (1) Integrating multi-temporal satellite data (Formosat-5 and Sentinel-2) as inputs for AI models. (2) A manual SLIP based landslide detection approach is applied to automate the labelling process in the training stage of AI model. (3) An executable AI based landslide detection model with trained database is carried out and corresponding API connection is established. (4) A test procedure of shared model including the submission of spatial-temporal criterion and retrieving the final detected landslide area is demonstrated.

Hardware description languages are not the easiest way to implement hardware designs for on-board hyperspectral image processing, therefore alternatives like HLS have become very popular to speed up the task. Nevertheless, HLS solutions are still behind RTL designs in terms of performance. Aimed at filling the gap between traditional HDLs and HLS, this work proposes a new hardware description language, based on simple predefined modules which perform well-defined tasks. The modules are meant to be chained together, defining multi-path pipelines. In contrast to traditional architectures, there is no standalone control unit, since the logic is embedded into the data flow.

Onion is a high-value crop that is highly susceptible to price fluctuations in the Philippines due to weather conditions, global political situations, and other factors. Accurate mapping and yield monitoring is crucial for managing these fluctuations and ensuring stable supply chains. Unlike multispectral satellite images, hyperspectral images offer higher spectral resolution that enable it to differentiate subtle variations in the spectral signatures of onions compared to other crops. Thus, this study explores the effectiveness of PRISMA hyperspectral imagery for mapping onion fields through two distinct methodologies: K-means unsupervised classification and Linear Spectral Unmixing (LSU). The PRISMA image, captured on February 4, 2024, covers the area of Bongabon, Nueva Ecija, known as the onion capital of the Philippines, and its surrounding municipalities. The Level 2D product was denoised using Minimum Noise Fraction (MNF) by Forward MNF followed by Inverse MNF. The dimensionality of the image was then reduced using Principal Component Analysis (PCA). Three sets of data inputs - PC 1-2, PC 1-4, and 175 PRISMA bands - were classified using K-means. Separately, linear spectral unmixing was performed using four representative spectral signatures for each class - onion, rice, and soil – extracted from denoised PRISMA using known field locations. By comparing the outcomes of these methodologies, this research evaluates their accuracies in delineating the onions, with LSU providing more precise quantification of onion extent. The results highlight the potential of hyperspectral remote sensing in precision farming and in effective mapping and monitoring of onion yields to help mitigate market volatility.

High-quality reference images are crucial for empirical model-based atmospheric corrections. The Taiwan Space Agency (TASA) has developed an approach that uses Surface Reflectance (SR) from Sentinel-2 as a reference for these corrections. To enhance efficiency, the selection and preparation of reference images must be automated. Therefore, a procedure for optimal reference image selection has been developed. This procedure includes three main steps: First, Setting the search criteria based on input images, such as acquisition date and geographic locations. Second, using remote servers to search for all available reference images within the given period and calculating cloud cover over land. Third, retrieving the top three cloudless reference images as candidates for atmospheric corrections. After applying atmospheric corrections, the result with the most Pseudo-Invariant Features (PIFs) will be selected as the final output.

This paper presents a comprehensive radiometric calibration approach for THEOS-1, a satellite owned by the Geo-Informatics and Space Technology Development Agency (GISTDA) in Thailand. Radiometric calibration is crucial for ensuring the accuracy and reliability of satellite imagery, particularly for applications such as environmental monitoring, disaster management, and land cover classification. In this study, we employ a combination of vicarious and cross-calibration methods to refine the radiometric calibration parameters of THEOS-1. The Taiwan Space Agency (TASA) has collaborated with GISTDA to develop and validate the radiometric calibration procedures for THEOS-1. The vicarious calibration method involves utilizing ground-based measurements of known radiance targets to establish a relationship between the satellite sensor's digital numbers and the actual radiance values. We leverage the extensive work with the support of onsite sun photometer measurements and MODIS atmospheric data to obtain high-quality atmospheric data, which is essential for accurately characterizing the radiative transfer processes in the Earth's atmosphere. Furthermore, we incorporate a cross-calibration step to enhance the accuracy of the radiometric calibration. This involves comparing the radiance measurements of THEOS-1 with those of well-calibrated reference sensors on other satellite platforms. By leveraging the radiometric consistency between multiple sensors, we can identify and correct systematic biases, thereby improving the overall calibration accuracy of THEOS-1. The proposed calibration approach is implemented and validated using a series of observational data acquired by THEOS-1 over various calibration sites with diverse surface types and atmospheric conditions. Preliminary results demonstrate significant improvements in the radiometric accuracy of THEOS-1 imagery, thereby enhancing its utility for a wide range of Earth observation applications. This research demonstrates that TASA and GISTDA have successfully developed a robust procedure for the radiometric calibration of THEOS-1', making it ready for routine calibration. The collaboration between TASA and GISTDA has resulted in a well-established calibration protocol that ensures the high quality of THEOS-1 data. In summary, this paper contributes to the ongoing efforts to enhance the radiometric calibration of THEOS-1 by leveraging both vicarious and cross-calibration methods. The calibrated imagery generated through this approach holds great promise for advancing scientific research, environmental monitoring, and societal applications reliant on satellite-based Earth observation data.

Land Surface Temperature (LST) is important data for various fields, especially for monitoring global warming. Landsat-8 satellite imagery provides thermal data with a spatial resolution of 30m (resampled from 100m) as the main data for LST retrieval. This research aims to compare several LST retrieval algorithms such as LST retrieval using band ten, Single Channel Model (SCM), Qin’s Split-Window Algorithm (Q-SWA), Sobrino’s Split-Window Algorithm (S-SWA), and LST from the analysis ready data of Landsat-8 Level two. The study focuses on Dallas, Texas, and surrounding areas in March 2023. We collected air temperature data from 20 U.S. Environmental Protection Agency (EPA) stations for indirect validation. Based on the results, the Q-SWA method outperformed other retrieval algorithms with Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) were 0.683°C and 0.842°C, respectively. Given that Landsat-8’s thermal band data is resampled from 100m to 30m, this research also explores enhancing the retrieved LST using Deep Neural Network Regression (DNNR). The best retrieved LST from the Q-SWA method served as the target data while several bands and spectral indices from Landsat-8 were used as the input for DNNR model. Due to the large scale of data, we randomly selected ten million pixels and divided into 80% of training and 20% of testing data. The DNNR model achieved MAE of 1.022°C on the testing dataset. The enhanced LST from the DNNR model was also validated with the same air temperature validation data and achieved the MAE score of 1.037°C. Based on the visual comparison result, the DNNR model successfully enhanced the retrieved LST by providing more detailed results at the same 30m resolution and showing promising performance based on error metrics. This finding suggests the potential for using deep learning regression in LST downscaling to achieve better spatial resolution.

Accurate and precise monitoring of coastal environments allows for the better preservation of their biodiversity. This study applies multispectral imaging, Unmanned Aerial Vehicles (UAVs), and supervised and unsupervised techniques to characterize a coastal area in Batangas, Philippines. Multispectral image data was gathered using a DJI Mavic 3M drone. Afterwards, vegetation maps using NDVI, GNDVI, NDRE, and LCI were generated. Regions of the image were then clustered using the k-means clustering algorithm to define habitats in the area of study. These clusters were then used to train supervised machine-learning algorithms for pixel-based image classification. After classifying the entire image with these models, the identified habitats were characterized based on their associated vegetation index measurements. It was found that aquatic areas of the image possessed scores associated with healthy and photosynthetically active water.

To enhance the spatial resolution of VRE bands of Sentinel-2 image and perform a jointed analysis with Pléiades-1 images, the renowned small-data learning theory, convex/deep (CODE), is employed, forming a completely unsupervised high-quality synthesis system with fast closed-form implementation. More specifically, we formulate this synthesis problem as a satellite image fusion problem (i.e., Sentinel-2 and Pléiades-1 images). Initially, both images are fused by the proposed unsupervised Deep Residual Convolution Network (DRCN). The remarkable AI property, Deep Image Prior (DIP), enables DRCN to generate a preliminary rough solution entailing feasible spectral characteristics. Subsequently the convex Q-quadratic norm is leveraged to bridge the convex optimization and deep learning. Eventually, a powerful convex solver, the Alternating Direction Method of Multipliers (ADMM), is employed for solving the convex problem; this is the essence of CODE theory. Owing to the superiority of small-data learning technology, the high-quality synthesis of 4m/2.5m image products in multiple VRE bands for Sentinel-2 data can be acquired in a completely unsupervised manner.

Traditional multispectral Color Filter solutions require a costly, complex wafer coatings process on a single glass, in partnership with TASA, Liscotech and TASA co-developed a mechanical architecture, named Multi-Band-in-One (MBIO) which defines multi-trench design, supporting individual off-the-shelf color filters, reducing production costs and accelerating lead times. This MBIO architecture has been proven successfully in Liscotech RSI3000, equipped with R/G/B/PAN/NIR MBIO filters and the AMS CMV12000 sensor, making it ideal for detailed EO. The SWIR1000, featuring Sony’s IMX991 SWIR sensor, excels in shortwave infrared (SWIR) imaging for weather and environmental monitoring. Both models leverage AMD UltraScale FPGA, enabling real-time preprocessing like AG, AWB, and Edge AI on the sensor side. Platforms have undergone rigorous testing, including radiation, vacuum thermal cycling, and vibration, ensuring reliability in extreme space conditions.

Meta publicly released the Segment Anything Model (SAM), an AI model specialized for image segmentation purposes, in 2023. Then, based on this scheme, there has been a lot of research on applying it to object recognition and extraction problems in various applications. There is also active research on derivation techniques to improve the performance and accuracy of SAM. SAM supports zero-shot or few-shot generalization, which does not require the collection of its segmentation data and does not require the model to be labeled for the use case. This research is a case study of SAM application using a fusion of high-resolution optical and SAR imagery as input. The images used in this study are KOMPSAT-3/3A optical images and KOMPSAT-5 Synthetic Aperture Radar (SAR) images, and the purpose of the application is water object extraction. The input images are fused by applying the wavelet technique. The study area is Boryeong City in Chungcheongnam-do, west coast area in the Korean Peninsula. The results obtained by applying SAM to this data were compared with the digital map of the river boundary, lake, or reservoir of the National Geographic Information Center in Korea and the OpenStreetMap (OSM) water objects in the same study area for quantitative accuracy comparison analysis. The Weight mean Intersection over Union (WmIoU) technique was used as an accuracy evaluation method. The experimental results show a significant improvement in accuracy when fused images are applied compared to optical images. As a result of this study, we expect that SAM, specializing in natural object extraction for multi-sensor image data, will be developed in the future and can be applied to solve many current problems.

Remote sensing has been applied to land, water, ocean, and atmosphere monitoring in earth science for several decades. UAV-based remote sensing also gradually becomes an important research or tool for better understanding our living earth and environment considering less cost, flexibility of operation, and controllability of image acquisition. A mutlispetcral imager with four spectral bands is designed for ground and vegetation monitoring that installed on a hybrid Vertical Take-Off and Landing (VTOL) UAV. A RGB camera with an IR-cut filter and a monochrome camera with an IR range or red-edge range filter are included. Passive altitude controlled is adopted to reduce the power requirement.

Bushfires are key drivers of ecological processes, particularly in fire-prone regions like Australia, where they shape forest ecosystems and affect biodiversity. Monitoring landscape recovery dynamics after bushfires is crucial for understanding ecosystem resilience. This study focuses on monitoring post-bushfire landscape recovery dynamics using remote sensing data and cloud computing. Landsat imagery from 2007 to 2024 was analyzed via Google Earth Engine (GEE) to create cloud-free surface reflectance composites. The primary objective was to assess spectral recovery patterns across different burn severity classes in two areas of Southeast Victoria, Australia. The study utilized Normalized Difference Vegetation Index (NDVI) and Normalized Burn Ratio (NBR) as spectral indices to evaluate the post-fire spectral recovery. Findings revealed that the NBR demonstrated a longer spectral recovery duration compared to NDVI, especially in areas severely affected by bushfires. The results show that remote sensing combined with spectral indices is an effective approach to detecting, mapping, and understanding the recovery processes in post-bushfire landscapes. The study highlights the utility of remote sensing technologies in environmental monitoring and emphasizes the need for further research to refine these methodologies and address limitations, such as the inability to capture sub-canopy dynamics and the effects of topography on recovery. This research provides valuable insights for improving environmental management strategies and enhancing the understanding of landscape recovery following severe fire disturbances.