This research article presents a comprehensive investigation of leaf area index (LAI) estimation using Sentinel-1 synthetic aperture radar (SAR) and Sentinel-2 Optical L2A datasets for the wheat crop. The water cloud model (WCM) and PROSAIL radiative transfer models (RTMs) are used to estimate LAI from SAR and optical data, respectively. To model the surface backscattering in WCM, the integral equation model (IEM) at VV and VH polarizations is used with the Gaussian correlation function. The results demonstrate that LAI derived from SAR at VH polarization ($R^{2}=0.72, \ \text {RMSE}= 0.60~\text {m}^{2}\text {m}^{-2}$) exhibits superior accuracy compared with optical LAI ($R^{2}=0.70,\ \text {RMSE}=0.82~\text {m}^{2}\text {m}^{-2}$). A fusion approach incorporating deep learning, principal component analysis (PCA), and nonlinear regression techniques is applied to fuse the SAR and optical datasets to further enhance LAI estimation accuracy. The accuracy of these estimations is tested against the ground-truth LAI taken at different locations. Among the fusion methods tested, deep learning emerges as the most effective and accurate approach ($R^{2}=0.91,\ \text {RMSE}= 0.38~ \text {m}^{2}\text {m}^{-2}$). This study provides valuable insights into the estimation of LAI using multisource remote sensing data and highlights the potential of deep learning for improved accuracy in fusion applications.

Monitoring and management of threatened, and medicinally important floral species in the complicated terrain are challenging and thus need advanced technological development. In this study, hyperspectral bands selected from radiometer data were used to identify sensitive bands for the marker compound at the genus level which could be upscaled to a larger scale for precise monitoring. Rhododendron is one of the keystone altitudinal species which is used for such purposes. Rhododendron arboreum flower is reported to be effective as a diuretic, choleretic, chronic diarrhoea, and anti-irritable bowel syndrome therapy. Its leaves are often used as a substitute for another famous plant drug Rohitaka. R. Arboreum, a tree species whose hyperspectral data were acquired in the complicated terrain in different regions of the Western Himalayas by sampling. Secondary metabolite (marker biomedicinal compound) was measured using chromatographic techniques and corresponding wavebands on their hyperspectral signature were identified using several sophisticated filtering techniques to denoise and select the desired feature with feature selection algorithms. Two band combination was developed using these identified wavebands for the secondary metabolite concentration estimation (Quer.) which provided satisfactory results for the same.Since these compounds are metabolic products, multiple aspects such as edaphic, biochemical, and topographic factors need to be assessed in determining the variation in the flavonoid content accordingly. Along with sensor-based hyperspectral band combination information was tested together with statistical unfolding approaches to understanding the influencing factors for flavonoids in the montane region. The result implies that R. arboreum developed band combinations based on selected wavebands only to determine the desired secondary metabolite concentration. On spectra, its values do not coincide with any other primary metabolite found in the R. arboreum. Hence the hyperspectral data are suitable for the estimation of phytochemical compounds. This will help in precise long-term mapping and monitoring of this species at a spatial scale without reciprocating any lab-scale technique each time.

Summary Human activity recognition (HAR) has gained researcher's interest due to its increasing demand in automated monitoring applications. Development of efficient HAR algorithm is still an open research area due to the challenges like inter and intra‐class variations, diversity in lighting conditions, view point changes, and complex object motions. Convolutional neural network (CNN) based methods have achieved significant improvement in HAR. However, CNN implementations have drawback that it require a lot of computational resources due to the use of large number of learnable parameters. To overcome this drawback, we propose a simple and computationally efficient deep CNN architecture using multi‐layer information fusion for HAR. In this study, we explore the impact of information fusion at intermediate layers of the network, as each convolutional layer of the network hierarchically extracts information at different level of abstraction of the objects from the video frames. In this work, first we designed a simple and computationally efficient deep CNN architecture and then we introduce a feature fusion strategy to integrate the complementary information of intermediate layers to the layer of the proposed CNN architecture. The proposed architecture is fine‐tuned and trained from scratch with raw RGB data. Softmax classifier is used at the last layer of network for activity classification. Benefits of the proposed architecture over standard deep learning architectures is it's high computational efficiency and reduced requirement of computational resources. To prove the effectiveness of the proposed method, we performed several extensive experiments on publically available datasets. The experimental results of the proposed method have demonstrated its superiority over other existing state‐of‐the‐art methods.

The present study carried out to parameterize the single channel soil moisture active passive (SMAP) passive soil moisture (SM) retrieval algorithm, over Indian conditions. The moderate resolution imaging spectroradiometer (MODIS) data products and soil texture data were used for an improved parameterization of the algorithm. The bias correction was applied to the MODIS leaf area index (LAI) for accurate computation of vegetation optical depth. The necessary vegetation and roughness parameter were calibrated through minimization of the error between model retrieved and ground measured SM. The value of root mean square error (RMSE) for retrieved SM was found as$0.059\,\,m^{3}m^{-3}$with bias and correlation coefficients of$0.036\,\,m^{3}m^{-3}$and 0.724 for ascending overpass, respectively, while a lower value was recorded (RMSE =$0.059\,\,m^{3}m^{-3}$, bias =$0.024\,\,m^{3}m^{-3}$, and correlation coefficients = 0.752) for descending overpass. The same method is also implemented on two other test sites in different regions of India to check the model robustness, which indicates that the current parameterization provides a better estimate of SM over croplands in India. The overall performance of new parameterized model is found as (RMSE = 0.052 and bias = 0.034) for ascending and descending (RMSE = 0.048 and bias = 0.026) satellite overpasses for all the three test sites. Additionally, the intercomparing of various operational SM products SMAP SM (L2_SM_P), Soil Moisture and Ocean Salinity (SMOS) SM (SMOS_L3_SM), and SMOS-IC data products was carried out with the SAC-ISRO PAN India SM network, which showed a significant RMSE, dry and wet biases over all three test sites as compared to the developed improved parameterized algorithm.

The main objective of the study was to validate the Sentinel-2 Level 2 Prototype Processor (SNAP-SL2P) derived LAI with ground observations and to verify consistency of global LAI products with ground observed LAI and upscaled validated LAI products. In present study, decametric Sentinel-2 LAI product was retrieved by the SL2P and validated using field measured LAI. Validated LAI product was upscaled at Copernicus (Sentinel-3/OLCI, PROBA-V), MODIS (combined terra and aqua) and VIIRS spatial scales for intercomparison with ground observed LAI and global LAI products, was presented as a way forward to validate hectometric LAI products.

The crops’ biophysical parameters play an important role in balancing the land surface energy fluxes and are needed in crop simulation modeling, evapotranspiration, etc. The vegetation parameters’ retrieval using microwave scattering model, mainly affected by the heterogeneous distribution of land targets, hampers an accurate retrieval of soil-vegetation parameters in microwave remote-sensing algorithms. To minimize the errors in biophysical parameters’ retrieval, the synergetic approach of modified water cloud model (MWCM) and modified soil scattering model (MSSM) was attempted to retrieve the leaf area index (LAI) of wheat and barley crops. Due to the spatiotemporal resolution of Sentinel-1A synthetic aperture radar (SAR) mission, it could be more sensitive to vegetation condition and the retrieval accuracy than optical/IR satellites. The nonlinear least square optimization algorithms were used for the parameterization of modified scattering model. The lookup table (LUT)-based inversion algorithm was applied to compute the LAI values through the modified scattering models. The statistical analysis was performed to assess the model efficiency. In case of forward modeling, the highest${R} ^{{2}} =0.96$and low RMSE = 0.20 dB were computed between the modeled$\sigma ^{\mathbf {0}}$(dB) and SAR-derived$\sigma ^{\mathbf {0}}$(dB) using vegetation descriptor (${V}$) = LAI. On the other hand, for inverse modeling, the LAI values obtained were more accurate at VV polarization (${R} ^{\mathbf {2}} =0.94$and RMSE$=0.124~\text{m}^{\mathbf {2}}/\text{m}^{\mathbf {2}}$), when compared with thein situdata. The overall product was also compared with the Project for On-Board Autonomy-Vegetation (PROBA-V) and Moderate Resolution Imaging Spectroradiometer (MODIS)-LAI to check the robustness of the approach.

The use of microwave observations in the low-frequency range is a complementary tool for mapping surface soil moisture. The L-band (1–2 GHz) region is a lower frequency band of microwave radiations, and currently, only two satellite soil moisture data products are available in the L-band frequency range through the satellites Soil Moisture Active Passive (SMAP) and Soil Moisture Ocean Salinity (SMOS). Both these have almost the same spatial resolution around 36–40 km. Although SMAP also provides an enhanced soil moisture product at 9 km, still finer scale information of soil moisture is required. The present study tries to enhance coarse-scale soil moisture by incorporating the single-channel algorithm (SCA). The results obtained by SCA were used as the inputs for the downscaling algorithms instead of directly using the satellite soil moisture product. Through this study, we implement and compare three approaches: approximation of thermal inertia (ATI), triangle, and dispatch methods. The results illustrated that the downscaling algorithms perform better with the estimated SMAP soil moisture through SCA in comparison to direct SMAP soil moisture, and the thermal inertia-based approach is the best performing among the three methods.

This paper demonstrates machine learning based approach to retrieve soil moisture (SM) and its validation over India using CYGNSS data. CYGNSS mission is mainly designed and dedicated for monitoring the tropical cyclones over ocean.However, recent developments has highlighted the potential of GNSS-Reflectometry for land applications, specially for SM with high spatio-temporal frequency over traditional satellite data sets. It can be directly utilized to retrieve SM as complementary data to fill the spatial and temporal gaps in satellite microwave radiometer derived SM, like from SMAP and SMOS mission to meet the requirements of high spatial and temporal frequency data sets for agricultural applications. In this work, we developed an Artificial Neural Network (ANN) framework to derive SM and validated at selected agricultural sites over India. SMAP derived vegetation and roughness parameters were also used as inputs for training of ANN model to add the effect of vegetation and roughness. Detailed spatial and temporal correlation analyses of CYGNSS SM were performed to test the proposed ANN model using SMAP SM and in-situ observations from hydra probe station data from 2018 to 2019. It was observed from temporal correlation analysis that CYGNSS and SMAP SM follow a good trend with high correlation using in-situ data. Spatial correlation also shows high correlation with Pearson correlation coefficient of 0.69 and RMSD of 0.057 m3/m3during pre-monsoon and 0.65 and 0.053 m3/m3in post monsoon periods, respectively.

The leaf area index (LAI) is a crucial parameter that governs the physical and biophysical processes of plant canopies and acts as an input variable in land surface and soil moisture modeling. The ScatSat-1 is the latest microwave Ku-band scatterometer mission of Indian Space Research Organization (ISRO), provides data at a higher temporal and spatial resolution for various applications. Due to its all-weather operational capability, it could be used as an alternative to the optical/IR sensors for the LAI estimation. In the technical literature domain, no testing has been done to estimate the LAI using ScatSat-1 scatterometer data. Therefore, the objective of this study is to retrieve the LAI using the ScatSat-1 backscattering by modifications of two different models viz. water cloud model (WCM) and the recently developed Oveisgharan et al. model and compared against the PROBA-V, MODIS, and ground-based LAI products. To assess the performance of these models, coefficient of determination (R2), root-mean-squared error (RMSE) and bias are computed. For Oveisgharan et al., the values of R2, RMSE and bias were obtained as 0.87, 0.57 m2m-2, and 0.05 m2m-2respectively, whereas for WCM model, the values were found as 0.82, 0.67 m2m-2, and 0.32 m2m-2respectively. This investigation showed that the modifications in Oveisgharan et al. model provide marginally better results in the retrieval of LAI using ScatSat-1 data than the WCM model. The models' limitation may be less serious for crop management studies because the majority of crops attains its maturity at LAI values less than 6 m2/m2.

Drought is a recurrent phenomenon in the semiarid regions of India that significantly affects the regional social, economic, and environmental conditions. Drought monitoring and assessment are challenging especially for regions that have sparse or very limited rain gauge observation networks. In this study, a comparative analysis was performed between satellite precipitation products including Tropical Rainfall Measuring Mission (TRMM-3B42), Climate Hazards Group InfraRed Precipitation (CHIRP) and ground-measured Indian Meteorological Department (IMD) precipitation data over the Bundelkhand region of Uttar Pradesh, India to assess the meteorological drought in this region. The district wise study was done to determine the regional differences among these satellite precipitation products and to statistically verify their performance in estimating the degree and spatial pattern over the study area. The Standardized Precipitation Index (SPI) was computed using the open-source DRINC software. The IMD derived SPI was found more correlated with TRMM compared to CHIRP for SPI-1, 3 and 12 with r values 0.74, 0.81 and 0.65 respectively. However, SPI-6 shows low positive and negative and correlation for both TRMM and CHIRP data. The current case study highlights the outperformance of TRMM data that enables real-time drought assessment owing to better accuracy and higher spatiotemporal resolution.

With the growth of various image-capturing devices, image acquisition is no longer a difficult task. As this technology is flourishing, various types of complex images are being produced. In order to access a large number of images stored in database easily, the images must be properly organized. Field of image retrieval attempts to solve this problem. As the complex images are being produced, processing them using single-resolution techniques is not sufficient as these images may contain varying levels of details. This paper proposes a novel multiresolution descriptor, local binary curvelet co-occurrence pattern, to achieve the task of content-based image retrieval. Curvelet transform of grayscale image is computed followed by computation of local binary pattern of resulting curvelet coefficients. Finally, feature vector is constructed using grey-level co-occurrence matrix which is matched with the feature vector of database images. The proposed descriptor combines the properties of local pattern and multiresolution technique of curvelet transform, and efficiently covers curvilinear and geometrical structures present in the image. Performance of the proposed method is measured in terms of precision and recall and is tested on five benchmark datasets consisting of natural images. The proposed method has been compared with single and multiresolution techniques as well as with some of the other state-of-the-art image retrieval methods. The experimental results clearly demonstrate that the proposed method produces high retrieval accuracy and outperforms other techniques in terms of precision and recall.

Vegetation water content (VWC) is an important component of microwave soil moisture retrieval algorithms. This paper aims to estimate VWC using L band active and passive radar/radiometer datasets obtained from a NASA ground-based Soil Moisture Active Passive (SMAP) simulator known as ComRAD (Combined Radar/Radiometer). Several approaches to derive vegetation information from radar and radiometer data such as HH, HV, VV, Microwave Polarization Difference Index (MPDI), HH/VV ratio, HV/(HH+VV), HV/(HH+HV+VV) and Radar Vegetation Index (RVI) are tested for VWC estimation through a generalized linear model (GLM). The overall analysis indicates that HV radar backscattering could be used for VWC content estimation with highest performance followed by HH, VV, MPDI, RVI, and other ratios.

NASA's Soil Moisture Active Passive (SMAP) mission is scheduled for launch in November 2014. In the prelaunch time frame, the SMAP team has focused on improving retrieval algorithms for the various SMAP baseline data products. The SMAP passive-only soil moisture product depends on accurate parameterization of the tau-omega model to achieve the required accuracy in soil moisture retrieval. During a field experiment (APEX12) conducted in the summer of 2012 under dry conditions in Maryland, the ComRAD truck-based SMAP simulator collected active/passive microwave time series data at the SMAP incident angle of 40° over corn and soybeans throughout the crop growth cycle. A similar experiment was conducted only over corn in 2002 under normal moist conditions. Data from these two experiments will be analyzed and compared to evaluate how changes in vegetation conditions throughout the growing season in both a drought and normal year can affect parameterizations in the tau-omega model for more accurate soil moisture retrieval.

BACKGROUND: Profile Hidden Markov Models (HMM) are statistical representations of protein families derived from patterns of sequence conservation in multiple alignments and have been used in identifying remote homologues with considerable success. These conservation patterns arise from fold specific signals, shared across multiple families, and function specific signals unique to the families. The availability of sequences pre-classified according to their function permits the use of negative training sequences to improve the specificity of the HMM, both by optimizing the threshold cutoff and by modifying emission probabilities to minimize the influence of fold-specific signals. A protocol to generate family specific HMMs is described that first constructs a profile HMM from an alignment of the family's sequences and then uses this model to identify sequences belonging to other classes that score above the default threshold (false positives). Ten-fold cross validation is used to optimise the discrimination threshold score for the model. The advent of fast multiple alignment methods enables the use of the profile alignments to align the true and false positive sequences, and the resulting alignments are used to modify the emission probabilities in the original model. RESULTS: The protocol, called HMM-ModE, was validated on a set of sequences belonging to six sub-families of the AGC family of kinases. These sequences have an average sequence similarity of 63% among the group though each sub-group has a different substrate specificity. The optimisation of discrimination threshold, by using negative sequences scored against the model improves specificity in test cases from an average of 21% to 98%. Further discrimination by the HMM after modifying model probabilities using negative training sequences is provided in a few cases, the average specificity rising to 99%. Similar improvements were obtained with a sample of G-Protein coupled receptors sub-classified with respect to their substrate specificity, though the average sequence identity across the sub-families is just 20.6%. The protocol is applied in a high-throughput classification exercise on protein kinases. CONCLUSION: The protocol has the potential to maximise the contributions of discriminating residues to classify proteins based on their molecular function, using pre-classified positive and negative sequence training data. The high specificity of the method, and increasing availability of pre-classified sequence data holds the potential for its application in sequence annotation.