Читать книгу The Digital Agricultural Revolution - Группа авторов - Страница 64

The present study conducted with an aim to test the ability of machine learning algorithm in integration with remote sensing in crop yield prediction of paddy and sugarcane crops at regional level. Crop-sensitive parameters extracted from high-resolution LANDSAT 8 OLI imageries are used as neural network model inputs. The FFBPNN models for crop yield were developed and calibrated in MATLAB environment. During training, the model perceptrons were trained with 75 of the 100 inputs up to 10,000 epochs with 1 to 10 hidden neurons. Statistical analysis revealed the reliability of the model in paddy yield estimation. However, slight under estimation of yield of the sugarcane crop indicates sensitivity of yield algorithms to crop input parameters. It was concluded that there is a high efficacy of using remote sensing images and NN models to generate accurate crop yield maps and also revealed significant superiority of neural network models over conventional methods.