Journal of the Indian Society of Remote Sensing最新文献

Prediction of Tropical cyclones (TCs), particularly intensity prediction, has always been challenging for climate researchers due to the complicated physical mechanisms in TC dynamics and the way it interacts with upper-ocean and atmospheric circulation. Furthermore, the available data set over the North Indian Ocean (NIO) is also very limited for Machine Learning (ML) model development. Here, we demonstrated a simple yet robust hybrid architecture leveraging a Convolutional Neural Network for automated prediction of the intensity of the cyclone based on IR satellite imagery of 2000–2022. The model comprises a binary classifier, a multiclass classifier, a YOLOv3 based cyclone detector and a regression module. The paper also highlights the discrepancy between the results of independent testing wherein training is done on 2000 to 2019 dataset and tested on 2020 to 2022 dataset, as well as the outcomes of a stratified train-test split performed over the entire dataset using a 70:15:15 ratio for training, validation and testing, respectively. The model is tuned for the NIO region with a binary classification accuracy score of 98.4% (± 0.003), multiclass classification accuracy of 63.83% (± 1.3) and RMSE of 16.2 (± 0.9) knots on stratified split. The results highlight the careful interpretation of the DL model’s performance when applied to time series problems. Additionally, it discusses the limitations stemming from the dataset's small size and the challenges posed by the 5 kt resolution of the best track intensity estimation from the Indian Meteorological Department (IMD). The internal representations learned by the model through feature maps analysis were studied, shedding light on the model’s decision-making process. The study underscores the need for further data accumulation and highlights avenues for enhancing model performance in the future.

The information generated about crop harvesting can aid several purposes, including the maximization of crop yield, minimizing crop losses, assessing quality deterioration and crop health, and studying phenology. This study aims to detect the harvesting cycle of Sugarcane-plant and ratoon and analyze the underlying trends. The agriculture domain makes use of remote sensing data extensively for applications like crop yield forecast, crop type mapping, monitoring crop patterns, etc. Sugarcane is cultivated in abundance in the Muzzafarnagar district of Uttar Pradesh, India. The two variants of sugarcane (ratoon and plant) are commonly grown in this region along with other crops like wheat, paddy, and oil seeds (sesame). To monitor the harvesting of the sugarcane crop fields, the phenology of the crop type (from germination to maturity stage) was considered as base temporal data from the DOVE sensor. The temporal Planetscope DOVE sensor base data with particular harvesting dates were used to map harvested fields of sugarcane ratoon and plants on a particular date. Modified soil adjusted vegetation index 2 (MSAVI2) and its variant class-based sensor independent modified soil adjusted vegetation index 2 (CBSI-MSAVI2) were tested to reduce spectral dimensionality and map the harvested fields on approximately a weekly basis. The harvested sugarcane ratoon and plant fields were successfully mapped using the innovative machine-learning approach with a Mean Membership Difference (MMD) value of about 0.01 and 0.02 respectively.

This study investigates the spatiotemporal oscillations of cirrus characteristics over the Indian subcontinent from 2013 to 2021 using Moderate Resolution Imaging Spectroradiometer observations. Analyzing Cirrus Fraction (CiF), Cirrus Reflectance (CiR), and radiative characteristics using the Clouds and the Earth’s Radiant Energy System data, the distinct spatial and seasonal fluctuations with respect to the regional precipitation characteristics is unveiled. Radiative characteristics demonstrate significant seasonal influences on Net Short-Wave Flux, Net Long Wave Flux, and Net Total Flux (NETF). Through linear regression analysis, a strong positive correlation is found between CiF, CiR and precipitation, indicating a robust linear relationship. Seasonal variations in cloud parameters and radiative characteristics are examined, revealing heightened Cloud Optical Thickness and Cloud Effective Radius during the South West monsoon season compared to other seasons. CiR and NETF are notably elevated during the Monsoon. These findings underscore the significant impact of the rain on cloud properties and energy flux dynamics over the Indian subcontinent.

Identifying soil erosion-prone zones in an ungauged river basin is crucial for devising and implementing timely soil protection measures to mitigate soil degradation and protect soil quality. Soil erosion damages the fragile ecosystem, decreases soil fertility, and reduces reservoir water storage, thereby impacting food production. The prime motive of the current research work is to assess and categorize on the basis of priority the sub-watersheds (SWs) susceptible to substantial soil erosion in the Ponnaniyar River basin (an ungauged river basin) based on the morphometric parameters that impact soil erosion. To achieve this research objective, four multi-criteria decision-making (MCDM) approaches based on the outranking approach and synthesis method are adopted to facilitate the decision-making process by considering an integrated and balanced assessment of multiple complex parameters for devising effective soil conservation measures to minimize soil erosion. Cartosat-1 digital elevation model (DEM) is employed to extract eighteen morphometric parameters under linear, shape, areal, relief and hypsometric aspects. The priority of SWs obtained by different MCDM techniques is evaluated using percentage of variation and intensity of variation. The outcomes show that the MABAC method is effective in prioritizing SWs with the least percentage of variation (59.61%) and intensity of variation (4.397). It is also found to be the best method for integration with the RSS method for determining SW priority with a root sum of squares value of 43. SW1 is identified to be highly vulnerable to soil erosion with a grade average value of 1.00 followed by SW2 (3.00), SW3 (3.25) and SW13 (5.00), requiring immediate implementation of watershed planning and management measures to control the extent of soil erosion and safeguard soil resources.

The extraction of crop information is one of the important research directions for precision agriculture remote sensing. Crop extraction is of great significance in crop refinement management, precision fertilization, growth monitoring and yield precision estimation. The karst mountainous areas in southern China are characterized by undulating terrain, broken cultivated land, scattered spatial distribution of tobacco planting plots, uneven growth of plants, and mixed planting of crops. As the flight height of UAVs increases, the area of tobacco planting plots decreases, and the texture features become increasingly blurred, which increases the difficulty of segmentation and affects the recognition accuracy. We conducted this study to explore whether the high-resolution sample datasets and the trained U-Net model are suitable for cross-level recognition. In this study, DJI Mavic 2 Pro was used to collect UAV RGB images with flight heights of 50 m, 60 m, 70 m and 90 m in complex habitats for extracting tobacco plants from the U-Net model. The results are as follows: (1) The precision of tobacco plant segmentation at different altitudes is 50 m > 60 m > 70 m > 90 m, and Kappa coefficient is 0.92, 0.89, 0.86 and 0.34; the pressure is 0.96, 0.94, 0.93 and 0.22; the recall is 0.91, 0.90, 0.86 and 0.24; and the IoU is 0.88, 0.85, 0.8 and 0.23, respectively; and the precision of complex background segmentation is: a small number of weeds > a large number of weeds, and the plot is flat > the plot is broken. (2) With increasing flight height, the precision of tobacco segmentation of the U-Net model gradually decreases. Compared with 50 m, the precision of the 60 m segmentation results is reduced by 0.03, 0.02, 0.01 and 0.03, and that of 70 m is reduced by 0.06, 0.03, 0.05 and 0.08. The precision of the 90 m segmentation results is reduced by 0.58, 0.74, 0.67 and 0.65. The flight heights of 50 m, 60 m and 70 m have good experimental results, but the precision of 90 m segmentation is poor. The precision is mainly affected by the two factors of floor height and light. This study verified the feasibility and reliability of the high-precision extraction of tobacco plants at different altitudes by U-Net in complex habitats and has a certain reference value for research on the methodology and technical system of the deep learning recognition of crops in complex habitats in karst mountains.

Air pollution as a result of desertification and dust transportation is one of the critical environmental challenges in the arid and semi-arid regions. Urmia Lake, the largest inland lake of Iran has lost most of its water over the past 2 decades. The lake bed is known as one of the aerosol pollution sources in the northwestern Iran. Although recent studies contributed to investigate the impacts of the drying up of Urmia Lake on the local and regional air quality, there is still a need to identify spatiotemporal aerosol pollution and dust generation sources in the study area. In this study, remote sensing techniques, fuzzy logic and Principal Component Analysis (PCA) were used to identify dust hot spots in the south and east parts of the Lake, where recent studies have highlighted the dramatic extent of salinization and desertification. Based on the results of this study, the lake's contribution to the local aerosol pollution declines with increasing distance from it. The results indicated that the potential of dust forming on the east side of the lake has increased, presenting a variety of challenges for inhabitants, including health and biological hazards. The fuzzy results have a high correlation with Electrical Conductivity (EC) (0.69), Aerosol Optical Depth (AOD) (0.46), and Leaf Area Index (0.45), respectively, while wind speed (0.22) and slope (0.24) have the lower correlation. The results of PCA indicate that AOD, Digital Elevation Model, and EC have the highest percentage in identifying dust generation sources among the effective parameters in determining dust production sources.

After oil spills occur in the ocean, oil pollutants usually appear in the form of oil emulsions under the influence of hydrodynamics. Hyperspectral remote sensing technology, which provides abundant spectral information of ground objects, has the potential of fine-grained classification on the types of oil emulsions. Aiming at the practical applications of oil emulsion extraction in hyperspectral images (HSIs), this study proposes a semi-supervised model for oil emulsion identification by integrating an image segmentation algorithm with a deep-learning-based classification model. In the proposed approach, the training data were filtered from HSI using an image segmentation algorithm, based on which a 1-dimensional convolutional neural network (1D-CNN) was trained to identify oil emulsions in the HSI. The model was tested on the HSIs of Deepwater Horizon oil spills obtained by AVIRIS. The overall accuracy and standard performance measurements of the proposed model are higher than 94% on the extracted dataset. The results indicated that the proposed model achieved similar detection results on sea water as the supervised model, and even higher accuracies on oil emulsion type identification. As a semi-supervised model, it also avoids the lengthy and time-consuming data labelling and has the potential for operational oil emulsions extraction and quantification.

This study aimed to propose an improved Gram–Schmidt adaptive (GSA) pansharpening method using the support vector regression (SVR) and Markov random field (MRF) models in the cases of high ratios between spatial resolutions of LRMS and PAN images. In the present study, the SVR model was used to model the nonlinear relationship between the original LRMS images and the corresponding downsampled PAN image, thereby aiming to obtain the intensity component (({mathbf{I}}_{L})) of the upsampled MS image. Then, the initial pansharpened HRMS image was generated from the GSA pansharpening method with ({mathbf{I}}_{L}) calculated by the SVR model, which is denoted as GSA–SVR in this study. Finally, the quality of the initial pansharpened image was further improved by using the MRF model, which is denoted as GSA–SVR–MRF. A performance comparison of the GSA–SVR–MRF method with competitive pansharpening techniques as well as the GSA–SVR method demonstrated its superiority in maintaining the spatial and spectral details of the PAN and original LRMS images. The GSA–SVR–MRF method was found to be the best in terms of most quality indices.

Air pollution is an important global environmental issue impacting public health across the world. Innovative satellite-based technology has revolutionized the monitoring of air pollution, enabling assessments on various scales with unprecedented accuracy and coverage. The study attempts to estimate the seasonal and spatial fluctuations of various gaseous pollutants using Sentinel-5P TROPOMI satellite images at the district level in Odisha. In order to comprehend the environmental impact of air pollution, an effort must be made to assess potential greenhouse gas (GHG) emissions and potential acidification levels in Odisha. Results showed that potential emissions of greenhouse gases vary regionally and range from 378.82 g CO₂ equivalent/m² to 386.22 g CO₂ equivalent/m², while potential acidification levels range from 0.008 g SO₂ equivalent/m² to 0.034 g SO₂ equivalent/m². The north-western (Jharsuguda, Sambalpur, Bargarh, Sonepur, and Sundargarh) and north-central (Angul, Dhenkanal, and Deogarh) regions of Odisha exhibit high potential emissions of greenhouse gases and levels of acidification. This is attributed to comparatively higher concentrations of various pollutants stemming from sources like industrial and vehicle emissions. Although the satellite-based study enabled us to characterise the relative air pollution across the state, it necessitated a number of air pollution monitoring stations for validation purposes. A future road map to address climate change and environmental protection may be developed with the aid of local officials and policymakers.

The velocity and volume of MultiSpectral (MS) remote sensing data have recently increased exponentially. In recent times, however, the absence of a spatial data cube to store analysis-ready data (ARD) products for the Indian sensors’ data delimits its ready use and depreciates its value. Establishing a framework for storing, managing, and providing online processing ARD products for different sensors is necessary. The current work proposes a framework to produce ARD products by radiometrically correcting the data using the 6 S atmospheric correction and Shepherd Diamond-based terrain correction method to provide normalised surface reflectance. The generated ARD product for LISS-III shows a good correlation with the Planet Lab’s surface reflectance ARD product and an excellent correlation with the SACRS2- a Scheme for Atmospheric Correction of ResourceSat-2 corrected product. A frequency-based geometric correction algorithm provides RMSE of less than half a pixel registration error compared to LANDSAT-8 OLI orthorectified imagery. Finally, A Spatial Data Cube (SDC) with CARD4L metadata standard stores the ARD products post ingestion. The paper explains the complete integrated software development with an end-to-end processing chain of LISS III, an Indian optical sensor data.