Classification of Wheat Seeds Using Neural Network Backpropagation Algorithm

Abstract

There are various types of wheat scattered in the world. Usually it takes a long time to recognize the type of wheat seed by manual method because wheat germ has a physical appearance that looks the same as others. One method that can be used is an Artificial Neural Network. In this study, the data used were secondary data which consisted of data from the variable physical characteristics of wheat germ. The types of wheat seeds that are classified are 3. The Artificial Neural Network architecture used in this study is 5. By comparing the 5 Artificial Neural Network architectures, it is concluded that the architecture consisting of 3 layers and 4 layers is more precise in the classification of wheat germ types. The accuracy obtained by the 2 Artificial Neural Network architectures is 90% and 90%, respectively.

Multisensor Feature Fusion Based Rolling Bearing Fault Diagnosis Method

“…(2) Forward propagation. The hidden layer features of the first AE h 1 is used as the input of the second AE for unsupervised training until all hidden layers are trained; (3) The backpropagation (BP) algorithm [32] is used for supervised fine-tuning to further optimize all the weights and biases; (4) The last hidden layer feature, h n , of the DAEN is fed into the Softmax classifier;…”