Journal of the Indian Society of Remote Sensing最新文献

To further improve the accuracy of extracting farmland spatial distribution information, this thesis proposes an improved SegFormer model for extracting farmland spatial distribution information from unmanned aerial vehicle images. This method first introduces Efficient Channel Attention to optimize each transformer block in the encoder. Then, input the output results of each optimized block into the introduced BiFPN layer for enhanced feature extraction, and input the weighted fused multi-level features from the encoder into the decoder. By aggregating multi-level features through the Multi Layer Perceptron, local and global attention are combined, and then further weighted feature fusion is achieved through BiFPN. Finally, tthe Squeeze Excitation and Efficient Channel Attention was proposed to enhance channel features and improve model performance. The experimental results indicate that the improved SegFormer model’s mean intersection over union and mean pixel accuracy were 96.91 SegFormer model, it has increased by 1.55 union and pixel accuracy for farmland is 98.42 than other semantic segmentation models, effectively extract the extraction accuracy of farmland edges and small farmland from drone images.

Ocean colour spectral observations play a significant contribution in mapping the earth marine resources through measurements with its inverted geo-physical/biophysical parameters. The retrieval of parameters from the basic sensor measurements highly depends on atmospheric scattering and absorption of light energy by its constituents. Hence the quantitative applications using these datasets are directly affected by the uncertainty in radiative transfer modeling towards atmospheric scattering and absorption and associated sensor degradation with time. Here authors presented an automation of radiometric calibration approach for ocean colour monitor of Oceansat-II (Jan 2017–Dec 2017) dataset through top of the atmosphere radiance simulation using a non-linear optimization technique. This algorithm also provides an alternative approach of calibrating the sensor vicariously through reduced dependency of systematic congruent in-situ measurements. Since Kavaratti in Lakshadweep, India is already a well-known site for calibrating the ocean colour sensors. The OCM cloud free images over this calibration site are utilized to perform its radiometric assessment for the year 2017 using radiative transfer model coupled with bio-optical model where the synchronous, relevant model inputs are simulated. The significant variations in the radiometric calibration coefficients were realized across the spectral bands 412 to 865 nm i.e. 5.5% to 11.8% change were recorded in the year 2017 followed by 2.65 to 5.23% change within a month of March respectively.

Pan-sharpening is very often employed in remote sensing to transform low-resolution multispectral (LMS) images into equivalent high-resolution multispectral images (HMS). Images resulting from pan-sharpening are sharper and more detailed that is resulted by improving spatial features of the multispectral image. One such approach of jointly processing LMS and Panchromatic images is discussed in the present study. The decision-level fusion suggested here involves choosing or combining details from numerous sources by taking decisions while analyzing features recovered from the input images. The proposed methodology is an amalgamation of principal component analysis used for separating spatial and spectral details of LMS, non-subsampled contourlet transform for feature analysis, and Jaccard similarity index and block activity measurement for localized decision-based fusion. The algorithm tries to provide an adaptive approach to address the trade-off between spectral and spatial resolution. Self-Organizing Maps based clustering technique is employed with the intension of grouping the image pixels into three categories soil, water and vegetation. The paper highlights the performance comparison of proposed method with various pixel-level fusion techniques ranging from techniques from Intensity, Hue and Saturation (IHS) transform to Neural Networks based pan-sharpening methods. This comparison is implemented using various reference and non-reference indicators along with Kolmogorov–Smirnov test. Additional analysis using Kolmogorov–Smirnov test is done to statistically analyze spectral degradation. The comparative analysis provides enough evidence that the suggested method yields fused images with enhanced edge details without forgoing the spectral features which was also evident from the mutual information obtained from clustered images. The resulting sharpened images tend to possess good spatial and spectral details that would simplify the automatic image analysis.

Building extraction from high-resolution images has been a fundamental task in the remote sensing field. It helps in monitoring natural disasters and developing urban areas. Encoder–Decoder based convolutional neural network (CNN) has provided a paradigm for automatic building extraction. However, extracting building information is difficult due to many reasons like diverse scales, complex background and variety of building structures. Moreover, achieving accurate boundary information remains challenging due to various impediments surrounding buildings. To deal with these challenges, in this article, we proposed a dual-branch model. One branch is the segmentation branch that includes an encoder–decoder framework (based on Attention-ResUNet architecture) combining residual unit and attention network, to generate the segmentation mask. The residual unit improves the ability to learn the deep and complex building features whereas the attention network focuses on the informative spatial information. In addition, a dilated module is positioned at the end of the decoder of Attention-ResUNet to capture the multiscale information. Another branch is the edge branch consisting of canny edge extraction, morphological operation and squeeze-excitation network, to improve the boundary information. The canny edge detection method extracts the edges of the buildings which is further enhanced through the morphological operation. In addition, a squeeze-excitation network is added for fine adjustment of generated feature maps. At the end, our proposed model integrates the segmentation mask obtained using the segmentation branch and boundary information generated by the edge branch to produce the refined segmentation mask. Experiments have been performed on the Massachusetts building dataset and the WHU-I building dataset. The performance of proposed model is compared with state-of-the-art models such as SegNet, DeepLabV3Plus, UNet, Attention-UNet, ResUNet and Attention-ResUNet. The results demonstrate that the proposed approach improves the performance for both the datasets. Hence, we can conclude that the proposed approach has a great potential in extracting multiscale information and enhancing the boundary information of buildings.

Satellite-based change detection involves comparing multi-temporal images to identify modifications in land cover features. This work investigates the application of a game theory classifier to enhance accuracy in medium-resolution multispectral remote sensing images. The proposed post-classification approach includes segmentation, feature extraction, classification, and image differencing to detect changes in multi-temporal images. To optimize multispectral images, land cover types are segmented using a proximal splitting algorithm. Boundary and texture features are then extracted using the Difference of Offset Gaussian Filter and Gray Level Co-occurrence Matrix. Principal Component Analysis is subsequently applied to reduce the dimensionality of the extracted features. Finally, the reduced features are classified using a game theory classifier, which effectively handles the uncertainty and variability inherent in non-smooth multispectral data. Experiments were conducted using Landsat datasets from the Hanoi and Balcoc regions, evaluating parameters such as misclassification rate, mean square error, color peak signal-to-noise ratio, and validity index. Quantitative analysis showed that the proposed approach achieved misclassification rates of 0.10 and 0.11 for dataset 1 and 2, respectively. Qualitatively, the results underscore the effectiveness of the extracted features in aiding the game theory classifier to discern subtle differences among feature classes.

The occurrence of heavy rains caused by cyclones has emerged as a significant factor leading to the occurrence of floods in the state of Andhra Pradesh, located in South India. The current investigation utilized a combination of GIS and AHP techniques to determine the flood-prone zonation of nine administrative units situated along the Vamsadhara River in the Srikakulam district. The analysis incorporated 16 parameters to identify the Flood Susceptible Zone (FSZ) and involved sensitivity analysis of the variables employed to enhance the reliability of the findings. The FSZ maps obtained were divided into five categories: very high, high, moderate, low, and very low. From the results, it was determined that 19% of the entire study area fell into the very high FSZ classification, while 34% were classified as high FSZ. Additionally, 29% of the area fell into the moderate FSZ category, followed by 14% in the low category, and 4% in the very low category. Among the 9 mandals selected for study, a majority of over 50% of the land area in Gara, Lakshminarsupeta, Narasannapeta, Polaki, and Sarubujjili faced susceptibility that varies from very high to highly susceptible to inundations. Overlay analysis of the water area on the FSZ map before and after a Cyclone demonstrates that the waterlogged regions predominantly coincide with the high and very high susceptibility categories. The results presented in the paper will provide valuable assistance to state and local officials by offering profound insights to support the implementation of effective strategies to reduce future risks.

Flooding stands as the most prevalent and financially burdensome natural disaster impacting nations worldwide. This study focuses on flood risk assessment within the Cuddalore taluk, aiming to leverage Geographic Information System (GIS)-based Analytic Hierarchy Process (AHP) techniques for analyzing flood hazards, vulnerabilities, and risks in the region. Seven key causal factors—elevation, slope, drainage density, river distance, rainfall, soil, and geology—were utilized to construct the flood hazard map. Results indicate that the taluk encompasses very low, low, moderate, high, and very high flood hazard zones, covering 7%, 22%, 34%, 25%, and 12% of its total area, respectively. Additionally, a flood vulnerability map was generated using five spatial layers: land use/cover, population density, distance to road, literacy rate, and population under the age of 6. Integration of the flood hazard and vulnerability maps facilitated the creation of a comprehensive flood risk map. The findings reveal that within the Cuddalore Taluk, zones classified as very low, low, moderate, high, and very high flood risk constitute 51%, 6%, 12%, 18%, and 12%, respectively. While the majority of the coastal region faces susceptibility to flooding within the very low, low, and moderate ranges, select areas are at risk of high and very high flooding. Disseminating flood hazard, vulnerability, and risk maps to relevant authorities is imperative for raising awareness regarding flood-prone locations. The coastal regions, along with adjacent areas, predominantly fall under the category of very high-risk zones, necessitating effective mitigation strategies. Specific locales such as Pillayarkuppam, Cuddalore, Tiruvandipuram, Kayalpattu, Nellikuppam, and Punjimangattuvalkkai demand focused efforts to mitigate high flood risks. Conversely, areas with very low and low flood risks, including Vadakuthu, Neyveli T.S., Sorathur, Panruti, Aierpali, and Pewndur, require preservation measures. Additionally, zones such as Arunam and Mettukuppam, exhibiting moderate flooding risks, warrant attention for preservation efforts in their immediate surroundings.

High-precision data on the Earth’s energy imbalance (EEI) require the creation of long-term fixed space platforms at a sufficient distance from the Earth. The Lunar Observatory (LO) is a single system of two identical special optical robotic telescopes installed along the equator at the opposite edges of the Moon, functioning sequentially as a single telescope. LO provides monitoring of the energy flux of the share of the total solar irradiance (TSI) reflected by the planet within the range of 0.2-4 micron and the outgoing intrinsic thermal radiation of the Earth within the ranges of 4–50 and 8–13 micron continuously during more than 94% of the lunar day. All these data will make it possible to calibrate and determine the dependence of the absolute value of the annual average EEI on cyclical TSI variations, which serves as a reliable indicator for reconstruction EEI variations for the total period of high-precision space TSI measurements since 1978. This will make it possible to reliably reveal the physical mechanisms of formation, reasons, and regularities of climate change on our planet. In the time free of the observations of the Earth LO will also produce a continuous all-sky survey: coordinate-photometric monitoring and study of near-Earth asteroids and comets, particularly moving from the side of the Sun, and also of exoplanets, supernovae and novae within the range of 0.2-2 micron and in its three individual broad bands.

The seismically active Shughnon district of the central-western part of the Gorno-Badakhshan Autonomous Region of Tajikistan nestled within the Pamir Mountain range, has experienced a significant number of seismic events of moderate to strong magnitude (> 4.0 to 6.8M_w) in the last decade. A deeper understanding and investigation of aftershock patterns and any potential seismic precursors is necessary for forecasting seismic hazards and its impact on human life, settlement and property. In this regard, utilization and correlation of night light disturbances data and night-time light emission anomalies against any pre- or post-seismic event offers a novel and unique approach to quantify the seismic impact on the mega cities especially situated in seismically active hilly and mountainous terrain. The study explores and examine the possible application of the Space borne night light imaging of Visible Infrared Imaging Radiometer Suite (VIIRS) as a potential tool for post-seismic impacts assessment and identification of any potential precursors or patterns of seismic activity and its impact on the mega city of Shughnon, Tajikistan. A retrospective analysis of VIIRS data cross-referencing with available historical seismic records of an eleven-year period (from 2012 to 2023) was evaluated and quantified by observing the variations in night light patterns and emission anomalies. About 15% reduction in night light brightness observed prior to three earthquakes, potentially linked to preemptive power grid shutdowns, infrastructure damage and population displacement. Post-earthquake imagery indicated a 60% decrease in lit areas and Recovery progress was quantified by a gradual 5% monthly increase in night light brightness, signaling restoration efforts.

Landslide is a common hazardous phenomenon in Bangladesh’s hilly areas, and Khagrachari is one of the regions that face frequent causalities due to landslide events. The present study has utilized the analytical hierarchy process (AHP) based multi-criteria evaluation techniques, frequency ratio (FR), modified frequency ratio (MFR), and information value method (IVM) approaches in the GIS environment to identify the landslide susceptible zones. The study uniquely employed 12 distinct parameters in this region to prepare the landslide susceptibility index (LSI) map of Khagrachari. The six unique LSI maps have been produced by three classification approaches, i.e., Quantile, Equal Interval, and Natural Break for decision matrix, and three different statistical modeling to compare the result. We found that the most susceptible zones of the Khagrachari district are Matiranga, Khagrachari Sadar, and Dighinala Upazila. The higher susceptibility has been primarily contributed by moderate-higher slope angle (14°–68°), high relative relief (176–601 m), geological structures, spares to moderate vegetation indices, and a high percentage of soil moisture (35–65%). Considering the classification approaches, around 9% of the area (~ 676 km²) is classified as a very high-hazard zone. In addition, we suggest that the MFR geospatial model has better prospects than IVM, AHP, and FR, as ~ 40% of the susceptible areas include more than 80% of the total landslide areas for the modified frequency ratio model. This study emphasizes the importance of implementing specific initiatives and activities to minimize landslide risks in Khagrachari. In addition, the present study installs the groundwork for future research to enhance geospatial modeling techniques and allows for comparisons with neighboring areas, thus expanding our knowledge of landslide susceptibility in the Chittagong Hill Tracts and adjacent regions of the Bengal Basin.