Artificial Neural Network Model for the Prediction of the Cotton Crop Leaf Area
A. M. Aboukarima *
Agricultural Engineering Research Institute, Agricultural Research Centre, Dokki, Giza, Egypt
H. A. Elsoury
Agricultural Engineering Research Institute, Agricultural Research Centre, Dokki, Giza, Egypt
M. Menyawi
Agricultural Engineering Research Institute, Agricultural Research Centre, Dokki, Giza, Egypt
*Author to whom correspondence should be addressed.
Abstract
Leaf area is an important indicator of crop growth and productivity. There are different instruments besides mathematical empirical models to estimate leaf area of crops, vegetables and fruits. This study investigates an Artificial Neural Network (ANN) model in prediction cotton leaf area. Best fitting results were obtained with 4 input nodes (leaf width, main lobe length, right lobe length and left lobe length), one hidden layer and one output layer (leaf area) as 4-6-1. ANN model performance was tested successfully to describe the relationship between measured and predicted cotton leaf area and coefficient of determination (R2) was 0.9232. The developed ANN model produced satisfied correlation between measured and predicted value and minimum inspection error. Thus, the model can be used in easy way for agronomists and plant scientists in cotton crop research.
Keywords: Artificial neural network, cotton, leaf area, modeling