Machine Learning-Based Remote Sensing Monitoring and Prediction Modeling of Bolboschoenusplaniculmis

Abstract

A multivariate regression approach is utilized to map the geographical dispersion of Above-ground biomass (AGB) of Bolboschoenusplaniculmis (B. planiculmis) using simultaneous moderate spatial resolution satellite images and field data. Bolboschoenusplaniculmis stands as an essential target species of plants for rehabilitation of damaged wetlands in the Momoge National Nature Reserve (MNNR), China's northeast. Initially the information is collected from Operational Land Imager (OLI) dataset, and then the collected data are fed to pre-processing segment. In pre-processing, Multi-window Savitzky-Golay Filtering (MWSGF) is used to preprocess the data to enhance image clarity. Then the preprocessed output is fed to Historical Information Passing Networks (HIPNs)is successfully used to predict the AGB of B. planiculmis. In general, Historical Information Passing Networks (HIPNs) the classifier does not articulate ways for optimizing parameters to guarantee accuracy AGB of B. planiculmi s prediction. Hence, proposed FLO enhances Historical Information Passing Networks (HIPNs), accurately predict the AGB of B. planiculmis. The weight parameter of the HIPN optimized with FLO for accurate prediction. The proposed RSPB-HIPN-FLO proposed is implemented on the Python working platform. The performance of proposed method examined utilizing performance metrics likes accuracy, precision; Root Mean Squared Error, Mean Absolute Error, and Correlation Coefficient were looked at. A suggested RSPB-HIPN-FLO approach contains 23.52%, 21.72%  and24.92%  higher accuracy; 23.52%, 22.72% and 21.92% higher Precision; and 24.58%, 21.71% and 22.90% lower and Root Mean Squared Errorlikened with current methods, like Monitoring invasive plant species using hyper spectral remote sensing data (PSHRS-ANN),Combining machine learning and remote sensing-integrated crop modeling for rice and soybean crop simulation (RSICS-DNN)and GOA-optimized deep learning for soybean yield estimation using multi-source remote sensing data(SYMRS-GOA) respectively.