Vinay Kumar Srivastava
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Vinay Kumar Srivastava has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
- Divyanshu Awasthi
Department of Electronics and Communication Engineering, GLA University, Mathura, India
25 shared publications - Priyank Khare
ECED, IIIT Ranchi
14 shared publications - Anurag Tiwari
Electronics and Communication Engineering Department, Motilal Nehru National Institute of Technology, Prayagraj, India
11 shared publications S. Tiwari
9 shared publications- Manoj Shukla
Harcourt Butler Tech Univ, Dept Elect Engn, Kanpur, Uttar Pradesh, India
6 shared publications Priyank Khare
6 shared publications
Latest Publications
- Open Access
- 18 June 2026
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 15 (6) , 275
Multisource Satellite Data-Driven Machine Learning Approach for Rice Yield Prediction
Estimation of rice crop yield at the village level is essential because village is the Insurance Unit (IU) for rice crop in many regions in India, and timely and accurate yield information at this scale supports timely and transparent claim settlements for farmers and supports local agricultural planning. To achieve this, a multi-source satellite data-based machine learning approach was used to estimate rice yield at the village level using optical and SAR data, climatic data and land surface model-derived parameters in Kakinada of Andhra Pradesh, India. The predictor dataset included seasonal cumulative rainfall, seasonal Normalized Difference Vegetation Index (NDVI)-Max, seasonal NDVI-Mean, seasonal Land Surface Water Index (LSWI)-Max, seasonal LSWI-Mean, season total Fraction of Absorbed Photosynthetically Active Radiation (fAPAR) and season total Root Zone Soil Moisture (RZSM), and season total backscatter of the Sentinel-1 VH polarization were used to represent crop greenness, moisture status, photosynthetic activity, soil water availability, canopy structure, and seasonal water supply. For model development and validation, village-level rice yield data from 2017 to 2023 was used, which was collected through Crop Cutting Experiment (CCE) at the maturity stage of Kharif season. In this study, four machine learning models such as Random Forest (RF), Support Vector Regression (SVR), Extreme Gradient Boosting (XGBoost), and Gradient Boosting (GB) were evaluated. The multi-source satellite data and yield data for the period 2017–2021 were used to train the models, which were independently tested on 2022 data and then applied to predict the rice yield in 2023. Leave-One-Year-Out (LOYO) cross-validation was also conducted on the 2017–2022 data to assess temporal robustness and generalization capability across years. Among the evaluated models, Random Forest exhibited the best overall performance. For the independent test year 2022, RF achieved an R2 of 0.465, RMSE of 415.34 kg ha−1, MAE of 322.22 kg ha−1, and MAPE of 10.36%. For the prediction year 2023, RF achieved improved accuracy with an R2 of 0.838, RMSE of 325.75 kg ha−1, MAE of 262.21 kg ha−1, and MAPE of 7.68%. Further, LOYO cross-validation also showed the robustness of RF, achieving the highest mean R2 of 0.702 and mean RMSE of 384.73 kg ha−1. The results illustrate that multi-source satellite data combined with machine learning can be a reliable and operationally useful tool in predicting village-level rice yield, which can be used for crop insurance claim settlement.
- 18 June 2026
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 15 (6) , 275
Open AccessMultisource Satellite Data-Driven Machine Learning Approach for Rice Yield Prediction
Estimation of rice crop yield at the village level is essential because village is the Insurance Unit (IU) for rice crop in many regions in India, and timely and accurate yield information at this scale supports timely and transparent claim settlements for farmers and supports local agricultural planning. To achieve this, a multi-source satellite data-based machine learning approach was used to estimate rice yield at the village level using optical and SAR data, climatic data and land surface model-derived parameters in Kakinada of Andhra Pradesh, India. The predictor dataset included seasonal cumulative rainfall, seasonal Normalized Difference Vegetation Index (NDVI)-Max, seasonal NDVI-Mean, seasonal Land Surface Water Index (LSWI)-Max, seasonal LSWI-Mean, season total Fraction of Absorbed Photosynthetically Active Radiation (fAPAR) and season total Root Zone Soil Moisture (RZSM), and season total backscatter of the Sentinel-1 VH polarization were used to represent crop greenness, moisture status, photosynthetic activity, soil water availability, canopy structure, and seasonal water supply. For model development and validation, village-level rice yield data from 2017 to 2023 was used, which was collected through Crop Cutting Experiment (CCE) at the maturity stage of Kharif season. In this study, four machine learning models such as Random Forest (RF), Support Vector Regression (SVR), Extreme Gradient Boosting (XGBoost), and Gradient Boosting (GB) were evaluated. The multi-source satellite data and yield data for the period 2017–2021 were used to train the models, which were independently tested on 2022 data and then applied to predict the rice yield in 2023. Leave-One-Year-Out (LOYO) cross-validation was also conducted on the 2017–2022 data to assess temporal robustness and generalization capability across years. Among the evaluated models, Random Forest exhibited the best overall performance. For the independent test year 2022, RF achieved an R2 of 0.465, RMSE of 415.34 kg ha−1, MAE of 322.22 kg ha−1, and MAPE of 10.36%. For the prediction year 2023, RF achieved improved accuracy with an R2 of 0.838, RMSE of 325.75 kg ha−1, MAE of 262.21 kg ha−1, and MAPE of 7.68%. Further, LOYO cross-validation also showed the robustness of RF, achieving the highest mean R2 of 0.702 and mean RMSE of 384.73 kg ha−1. The results illustrate that multi-source satellite data combined with machine learning can be a reliable and operationally useful tool in predicting village-level rice yield, which can be used for crop insurance claim settlement.
- 8 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE MultiMedia
- Vol. PP (1070986X) , 1-7
Skip-Mark: Class-based skip connected watermarking technique for the ownership protection of medical records
Telemedicine applications have gained popularity in recent times. Due to its usage, medical images are secured authentically, providing accurate diagnosis to the consumers. The proposed algorithm mainly focuses on developing a novel medical image watermarking technique in the deep learning (DL) environment. Class-wise perceptual Hash (Phash)-based watermarking is proposed to generate the watermark image from the host image itself. The watermarking process is implemented in DL environment, making the algorithm more robust and secure. Phash images are used for source authentication and classification of different classes of host images without knowing the original input image, taking into account the semantic information of each class. Our proposed technique offers a new paradigm in the integrity of clinical records. Authentication of the medical images is safeguarded robustly. Further, the proposed method is comprehensively analyzed for its performance. The percentage improvement in imperceptibility is 11.09%.
- 8 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE MultiMedia
- Vol. PP (1070986X) , 1-7
Skip-Mark: Class-based skip connected watermarking technique for the ownership protection of medical records
Telemedicine applications have gained popularity in recent times. Due to its usage, medical images are secured authentically, providing accurate diagnosis to the consumers. The proposed algorithm mainly focuses on developing a novel medical image watermarking technique in the deep learning (DL) environment. Class-wise perceptual Hash (Phash)-based watermarking is proposed to generate the watermark image from the host image itself. The watermarking process is implemented in DL environment, making the algorithm more robust and secure. Phash images are used for source authentication and classification of different classes of host images without knowing the original input image, taking into account the semantic information of each class. Our proposed technique offers a new paradigm in the integrity of clinical records. Authentication of the medical images is safeguarded robustly. Further, the proposed method is comprehensively analyzed for its performance. The percentage improvement in imperceptibility is 11.09%.
- 7 April 2026
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 99
XFaceMark: Explainable deep fake watermarking using YOLO, and random MRFO
- 7 April 2026
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 99
XFaceMark: Explainable deep fake watermarking using YOLO, and random MRFO
- RetractionOpen Access
- 19 March 2026
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 85 (4) , 296
Retraction Note: Image security enhancement to medical images by RDWT-DCT-Schur decomposition-based watermarking and its authentication using BRISK features
- 19 March 2026
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 85 (4) , 296
RetractionOpen AccessRetraction Note: Image security enhancement to medical images by RDWT-DCT-Schur decomposition-based watermarking and its authentication using BRISK features
- Open Access
- 10 February 2026
- Published by Institute of Central Computation and Knowledge Inc. in ICCK Transactions on Information Security and Cryptography
- Vol. 2 (1) , 43-54
Optimized Copyright Protection of Scale-Adaptive Saliency-Driven ROI of Medical Records with MSER-Based authentication
The medical information is always vulnerable to security violations. After the COVID-19 pandemic, the volume of medical information has exponentially increased and the major information is in the form of images. So, the security of this image information is crucial during the transfer using modern information and communication technologies from one place to another. Image watermarking is one of the methods to protect the copyright and integrity of medical records. The medical images consist of some area that has vital information regarding the disease and this area will be considered as the region of interest (ROI). Therefore, the protection of the copyright of this vital area is crucial for the effective diagnosis. The proposed technique utilizes a scale-adaptive anatomical saliency and a statistical information-guided ROI detection technique. The redundant wavelet transform (RDWT) and diagonalized Hessenberg decomposition (Diag-HD), along with the discrete cosine transform (DCT), are used for watermark embedding and extraction procedures. To get the optimized scaling weight, nature-inspired Salp Swarm Optimization (SSO) is used. Maximally stable extremal regions (MSER) are used for feature-based authentication. The simulation results analysis demonstrates the effectiveness in terms of visual similarity and robustness of the proposed technique. The experimental results obtained also highlight the superiority of the proposed technique over other existing methods in terms of imperceptibility and robustness.
- 10 February 2026
- Published by Institute of Central Computation and Knowledge Inc. in ICCK Transactions on Information Security and Cryptography
- Vol. 2 (1) , 43-54
Open AccessOptimized Copyright Protection of Scale-Adaptive Saliency-Driven ROI of Medical Records with MSER-Based authentication
The medical information is always vulnerable to security violations. After the COVID-19 pandemic, the volume of medical information has exponentially increased and the major information is in the form of images. So, the security of this image information is crucial during the transfer using modern information and communication technologies from one place to another. Image watermarking is one of the methods to protect the copyright and integrity of medical records. The medical images consist of some area that has vital information regarding the disease and this area will be considered as the region of interest (ROI). Therefore, the protection of the copyright of this vital area is crucial for the effective diagnosis. The proposed technique utilizes a scale-adaptive anatomical saliency and a statistical information-guided ROI detection technique. The redundant wavelet transform (RDWT) and diagonalized Hessenberg decomposition (Diag-HD), along with the discrete cosine transform (DCT), are used for watermark embedding and extraction procedures. To get the optimized scaling weight, nature-inspired Salp Swarm Optimization (SSO) is used. Maximally stable extremal regions (MSER) are used for feature-based authentication. The simulation results analysis demonstrates the effectiveness in terms of visual similarity and robustness of the proposed technique. The experimental results obtained also highlight the superiority of the proposed technique over other existing methods in terms of imperceptibility and robustness.
- 2 January 2026
- Published by Springer Nature
- p. 249-258
RPZMWMark: RDWT-PZM-Based Image Watermarking Scheme of DICOM Image
- 2 January 2026
- Published by Springer Nature
- p. 249-258
RPZMWMark: RDWT-PZM-Based Image Watermarking Scheme of DICOM Image
- 2 January 2026
- Published by Springer Nature
- p. 303-316
CFRV: Copyright Protection of Fundus Images and Real-Time Identity Verification for Smart Healthcare
- 2 January 2026
- Published by Springer Nature
- p. 303-316
CFRV: Copyright Protection of Fundus Images and Real-Time Identity Verification for Smart Healthcare
- 2 January 2026
- Published by Springer Nature
- p. 129-142
Fuzzymark: Fuzzy Based Watermarking for NIfTI Images Using RDWT-QR-RSVD
- 2 January 2026
- Published by Springer Nature
- p. 129-142
Fuzzymark: Fuzzy Based Watermarking for NIfTI Images Using RDWT-QR-RSVD
- 2 January 2026
- Published by Springer Nature
- p. 143-156
PZMSMark: Pseudo Zernike Moment Based Hiding Scheme for Medical Image in Schur Domain
- 2 January 2026
- Published by Springer Nature
- p. 143-156
PZMSMark: Pseudo Zernike Moment Based Hiding Scheme for Medical Image in Schur Domain
- 28 October 2025
- Published by Springer Nature
- p. 303-315
MECGWmark: Multiple ECG Watermarking Scheme in RDWT Domain Using Dual Decomposition for Smart Healthcare Application
- 28 October 2025
- Published by Springer Nature
- p. 303-315
MECGWmark: Multiple ECG Watermarking Scheme in RDWT Domain Using Dual Decomposition for Smart Healthcare Application