Image processing techniques and neural network models for predicting plant nitrate using aerial images

Image processing techniques were used to extract statistical and five different textural features of multi-spectral bands of aerial images. Two different neural network architectures (e.g. back propagation and radial basis function) were used to develop twenty different models to predict plant (corn crop) nitrate. These neural networks used extracted image features as their inputs. Five different performance criteria were used to evaluate the performance of these neural network models. Radial basis function model based on green vegetation index textural features provided the best performance with an average accuracy of 92.1%.