Journal of the Indian Society of Remote Sensing最新文献

Deserts are unique ecosystems that provides suitable habitats to many floral and faunal species and that are beneficial to human beings in many ways. Desert ecosystems are affected by several natural and anthropogenic factors, resulting in the degradation of ecosystem goods and services provided by them. Thus, there is a need to monitor them. Accordingly, the ecological status of 34 major non-polar deserts of the world have been monitored for a period of four decades. We have used (i) vegetation cover and NDVI (vegetation density/vigour) as indicators of ecological conditions, and (ii), long term rainfall and temperature patterns to monitor the extent and the effect of climatic variations. Among the 34 deserts, Taklimakan has consistently the lowest NDVI, while Tanami has the highest NDVI during the entire monitoring period. The Asian Kavir and Kharan deserts have the lowest vegetation cover; Tanami has the highest vegetation cover. Out of 34 deserts, Gobi, Kalahari, Margo, Mu Us, Simpson, Strzelecki, Taklimakan and Thar deserts have shown an increasing trend in vegetation cover. While, Chalbi, Patagonian and Sonoran deserts have shown a decreasing trend. Thar, Sechura and Sahara have shown an increasing trend in precipitation, while Namib has shown an opposite trend. 31 deserts have shown an increasing trend in the temperature. Present study is important as changes in the ecological conditions of the deserts have a profound impact on the land surface albedo, surface energy balance, regional climate, carbon sequestration, biodiversity, and global dust emissions.

With the advancement in satellite and Artificial Intelligence (AI), the increase in observation of the earth is increasing dramatically. With this development, the demand in the field of Remote Sensing (RS) is also growing rapidly. The spatial resolution and textural information of remote sensing images can be improved by introducing AI and Machine Learning (ML) technology. In the modern era of computer science, Deep Learning (DL) models are more familiar in the field of scene classification. This paper aims to develop a novel depth-wise CNN model to classify the RS images with low time effort during training with higher accuracy than the existing CNN model. For comparison, three typical CNN models of VGG16, VGG19, ResNet50 and RegNet are taken and tested on the RS datasets for classification. The experimented analysis demonstrates that the proposed classification model surpasses the existing classification models by producing higher accuracy in testing by taking a minimum time duration for training the RS datasets.

In this study, we estimated the spatial distribution of rock glacier velocities and elevation changes over the region of the Indian state of Himachal Pradesh. The rock glacier velocities are estimated by using the Differential Synthetic Aperture Radar Interferometric (DInSAR) technique. DInSAR is observed as accurate and optimum for rock glacier dynamic studies. From the 185 rock glaciers selected for this study, 127 of them are moving with a mean velocity of 35 cm/yr. (≈ 1 mm/day). However, none of these rock glaciers mean velocities exceed 100 cm/yr. The elevation change of rock glaciers is estimated using the DEM differencing technique. The DEM differencing with the SRTM C-band DEM and TanDEM-X DEM was employed to estimate the rock glaciers thickness change from 2000 to 2014. Among 185 glaciers, the mean thickness change is positive for 58 rock glaciers and negative for 127 rock glaciers. The elevation change of rock glaciers is ranging from − 19 m (thinning) to 10 m (mass gain) for 2000–2014 time period. The mean annual elevation change of these rock glaciers is observed as − 0.175 m. The study also compared the previous research work related to the ice glaciers velocity and thickness changes of this same region. Similar to the velocity, the elevation change of rock glaciers is smaller compared to the ice glaciers. In general, the thinning of the ice glaciers is associated with the ablation region, and mass gain with the accumulation region. Despite the significant evidence of thinning is observed mostly in the accumulation region for the rock glaciers. These elevation changes are correlated with the velocity measurements for the notable rock glaciers. Most of the rock glaciers observed high velocities in the accumulation region are also observed the thinning in the same region. The correlation of these two parameters might be associated with infiltration. However, the designated analysis of the rock glacier’s internal structure is expected to provide a correlation between the velocity and thickness change of the rock glaciers.

Soil salinity plays an important role in agriculture production and land degradation, especially in semi-arid and arid regions. Accurate prediction of soil salinity requires evaluating crop yield, native vegetation situations, and irrigation command area management. In this study, MLR (multiple linear regression), SVMs (support vector machines) and ANNs (artificial neural networks) models were employed by using Landsat-8 OLI and GIS (Geographical Information Systems) techniques for predicting soil salinity in northwest Iran. Soil salinity was measured at 92 points (in a depth of 0–20 cm). The vegetation and soil salinity spectral indices, extracted from Landsat-8 OLI, were employed as input data. The results of this study indicated that the best-developed model for predicting soil salinity was the SVM-based model with R² (0.874) and RPD (2.32) and the lowest RMSE (11.20 dS m⁻¹). Moreover, the performance of developed models under different vegetation coverage showed that the SVM-based model yielded the best result. It was concluded that the SVM-based model is reliable for quantifying soil salinization.

Thermal Discomfort Index has traditionally relied on parameters such as air temperature and relative humidity, obtained either from meteorological ground stations or through land-physical approaches estimated independently by satellites. These methods often fall short in adequately capturing both seasonal and detailed local spatial variations. This study addresses these limitations by establishing the Surface Thermal Discomfort Index (STDI), a composite of the Meteorological Discomfort Index (MDI) and the Discomfort Index over the land surface (DI_-Land). Focused on Ibu Kota Negara Nusantara (IKN) in East Kalimantan and neighboring cities, MDI is derived from reanalysis data (ERA5-Land), validated with ground station data, while DI_-Land is produced primarily from Landsat-8. An equal weighting factor was applied to MDI and DI_-Land for estimating STDI. Results indicate that STDI captures both seasonal and spatial variations, reaching peak level in May and October, and hitting a low point in July. The spatial distribution of STDI is influenced by landuse types. In 2023, IKN experienced an STDI of 26.2 °C, while Balikpapan and Samarinda recorded at 26.5 and 26.4 °C, respectively. Compared to previous study in Jakarta, IKN and neighboring cities’s STDI are higher up to 0.2 °C, remaining within the partially comfortable range in the tropics. Projecting IKN’s development until 2045, an annual MDI increase of 0.01 °C is anticipated. Moreover, a 4% rise in built-up areas is expected to elevate STDI by 0.1–0.2 °C. This study provides insights into the thermal discomfort status in cities across East Kalimantan, anticipating a gradual increase in discomfort levels during the development of IKN.

This study analysed the spatiotemporal changes in carbon stocks and Net Ecosystem Productivity (NEP) in Zoige County, Upper Yellow River, from 2000 to 2020 in response to China’s ecological civilization ideology and sustainable development. The carbon stock module of the InVEST model and carbon source/sink calculation formula were employed, and GeoDetector was used to analyze driving forces and spatial distributions. The findings were as follows: (1) The land use in Zoige County had undergone significant changes over the past two decades, characterized by a reduction in grassland area due to its conversion into woodland and peat wetland. (2) The carbon stock in Zoige County had consistently increased, accumulating 5.19 × 106 tons. (3) Zoige County had functioned as net ecosystem productivity (NEP) over the past two decades, with increasing trends, averaging 3.335 kg C/m². (4) The primary driving force behind changes in carbon stock and NEP were identified as ‘biological abundance’.

Wheat is a crucial global grain crop that plays a vital role in ensuring food security worldwide. The automatic and accurate counting of wheat ears is essential for assessing wheat yield. However, the detection accuracy is greatly affected by the complex background and small target size. To address these challenges and improve the performance, we propose an enhanced YOLOv5s method. In the backbone, we introduce the efficient channel attention (ECA) to enhance the feature extraction capability of the original C3 module. Additionally, we incorporate a small-scale detection layer in the neck and prediction stages. This modification expands the original three-scale feature detection (20 × 20, 40 × 40, and 80 × 80) to a four-scale feature detection (20 × 20, 40 × 40, 80 × 80, and 160 × 160), thereby enhancing the recognition accuracy of small targets. Experimental results demonstrate that our method achieves an Accuracy (Acc) of 93.97%, which represents a 2.94% improvement over the YOLOv5s. Additionally, our method has a mean absolute error (MAE) of 0.57, a reduction of 0.6 from the YOLOv5s. The Acc of the improved YOLOv5s approaches that of YOLOv7; however, the giga floating-point operations per second (GFLOPs) and inference speed of the enhanced YOLOv5s are significantly lower than those of YOLOv7. Across various phases of the wheat test dataset, the enhanced model demonstrated superior performance. As a result, the enhanced YOLOv5s enhances its suitability for challenging field conditions and offers a dependable technical framework for ear detection and wheat yield estimation.

Fog, a form of cloud in contact with the Earth’s surface, is one of the high-impact weather phenomena in northern India during the winter months. A new day-time fog detection scheme using the normalized difference snow index (NDSI) has been developed. The present analysis focuses on the detection of fog at high spatial resolution using data from the Resourcesat-2 AWiFS. The fog area detected is cross-validated with that detected using INSAT-3DR data at 1 km resolution using the same technique. The NDSI-based technique discussed here has shown a strong potential for fog detection during day-time. This study is also significant as a pre-launch sensitivity study for future GISAT with MX-VNIR, HyS-VNIR, HyS-SWIR, or similar other kinds of present-or-future sensors. Even though GISAT does not have a MX-SWIR channel, a combination of both MX-VNIR and HyS-SWIR with resampled spatial resolution may be useful for day-time fog detection using this technique.

Lunar exploration is pivotal in establishing a human presence on the Moon, and lunar crater detection plays a major role in this pursuit. The study is divided into two key phases: the creation of a specialized annotated dataset sourced from the Optical High-Resolution Camera on the Chandrayaan-2 satellite, and the evaluation of model performance using this dataset. Employing models such as FasterRCNN, YoloV5, and YoloV1, the investigation reveals the YoloV5 model’s superiority, achieving a precision of 92% and a recall of 83% for lunar crater detection. This finding constitutes a significant contribution to lunar exploration research.

Detection of clouds and shadows in remote sensing imagery is important due to its wide range of applications. There are a lot of applications in remote sensing images such as monitoring of the environment, change detection etc. It is an important and booming research area. Ineffective and inaccurate cloud and cloud shadow masking will cause undesirable effects on different task that can be performed by using remote sensing images. Because of high spectral conglomeration and the spectral and temperature discrepancy of the underlying surface the detection of clouds and associated shadows is not candid. In this paper, we propose CSDUNet a modified U-Net network for precise pixel-wise semantic segmentation of cloud and its associated shadow from optical remote sensing images. It uses an encoder network and a decoder network. This method concatenated feature maps at different scales. We have proposed a novel network for cloud detection, which extract features corresponding cloud and shadow at different scales from multilevel layers to generate sharp boundaries. Which will help to detect clouds in heterogeneous landscape, under complex underlying surfaces with varying geometry. Experimental analysis on the Landsat satellite dataset proves that the proposed CSDUNet achieves a dice coefficient of 95.05%. Our method got 95.93% precision, recall of 94.71% and Jaccard index of 97.29%. CSDUNet achieves accurate detection accuracy and surpass several traditional methods.