Affiliation:
1. Graphic Era deemed to be University
2. Graphic Era Hill University
Abstract
In this article, three different indices NDVI (Normalized Difference Vegetation Index), BNDVI (Blue Normalized Difference Vegetation Index) and GNDVI (Green Normalized Difference Vegetation Index) are used for the identification of wheat, mustard and sugarcane crop of Saharanpur district’s region of Uttar Pradesh. Sentinel 2B satellite images are collected from October 02, 2018 to April 15, 2019. These images are processed using Google Earth Engine. These sentinel images are used to generate NDVI, BNDVI and GNDVI images using GEE. These three different indices images are further processed using SNAP software and particular indices values for 210 different locations are calculated. The same process is used for calculating BNDVI and GNDVI values. ARIMA, LSTM and Prophet models are used to train the time series indices values (NDVI, BNDVI and GNDVI) of wheat, mustard and sugarcane crop. these models are used to analyse MSE (mean absolute percentage error) and RMSE values by considering various parameters. Using ARIMA Model, for wheat crop GNDVI indices shows minimum RMSE 0.020, For Sugarcane crop NDVI indices shows minimum RMSE 0.053, For Mustard crop GNDVI indices shows minimum RMSE 0.024. Using LSTM model, for wheat crop NDVI indices shows minimum RMSE 0.036, For Sugarcane crop BNDVI indices shows minimum RMSE 0.054, For Mustard crop GNDVI indices shows minimum RMSE 0.026. Using Prophet model, for wheat crop GNDVI indices shows minimum RMSE 0.055, For Sugarcane crop NDVI indices shows minimum RMSE 0.088, For Mustard crop GNDVI indices using Prophet model shows minimum RMSE 0.101.
Publisher
Environmental Research and Technology
Reference37 articles.
1. V. Avashia, S. Parihar, and A. Garg, “Evaluation of classification techniques for land use change mapping of Indian Cities,” Journal of the Indian Society of Remote Sensing, Vol. 48(6), pp. 877–908, 2020. [CrossRef]
2. P. Patil, V. Panpatil, and S. Kokate, “Crop prediction system using machine learning algorithms,” International Research Journal of Engineering and Technology, Vol. 7(2), pp. 748753, 2020.
3. B. E. Bunker, “Classification of satellite time series-derived land surface phenology focused on the northern fertile crescent,” [Dissertation thesis], University of Arkansas, 2013.
4. X. X. Zhou, Y.‑Y. Li, Y.‑K. Luo, Y.‑W. Sun, Y.‑J. Su, C.‑W. Tan, and Y.‑J. Liu, “Research on remote sensing classification of fruit trees based on Sentinel-2 multi-temporal imageries,” Scientific Reports, Vol. 12(1), 2022. [CrossRef]
5. L. Wang, Q. Dong, L. Yang, J. Gao, and J. Liu, “Crop classification based on a novel feature filtering and enhancement method,” Remote Sensing (Basel), Vol. 11(4), 2019. [CrossRef]