DETERMINATION OF THE NUMBER OF EARS IN IMAGES OF WHEAT FIELDS BY METHODS COMPUTER VISION
S. N. Tereshchenko, A. L. Osipov, E. D. Moiseeva
Novosibirsk State University of Economics and Management, Novosibirsk, Russia
Keywords: neural networks, artificial intelligence, wheat, ears, productivity, deep learning, augmentation, object detection
Abstract
The neural network technology is used to accurately calculate the number of wheat ears from photographs of a wheat field. The methods of deep learning of convolutional neural networks in interaction with the transfer learning methodology are used. With the help of the EfficientDet architecture, the neural network is trained, which allows the number of wheat ears to be determined from graphical images with accuracy of 0.88012 on a sample using the F1-score metric with a threshold value of 0.6 (the coincidence of the predicted markup and the actual one).
|
