Seed classification of three species of amaranth (Amaranthus spp.) using artificial neural network and canonical discriminant analysis

Abstract

The current study was conducted in 2013 to identify the seeds of three species of Amaranthus, Amaranthus viridis L., Amaranthus retroflexus L. and Amaranthus albus L., by using the artificial neural network (ANN) and canonical discriminant analysis (CDA) methods. To begin with, photographs were taken of the seeds and 13 morphological characteristics of each seed extracted as predictor variables. Backward regression was used to find the most influential variables and seven variables were derived. Thus, predictor variables were divided into two sets of 13 and seven morphological characteristics. The results showed that the recognition accuracy of the ANN made using 13 and seven predictor variables was 81.1 and 80.3%, respectively. Meanwhile, recognition accuracy of the CDA using the seven and 13 predictor variables was 74.0 and 75.7%, respectively. Therefore, in comparison to CDA, ANN showed higher identification accuracy; however, the difference was not statistically significant. Identification accuracy for A. retroflexus was higher using the CDA method than ANN, while the ANN method had higher recognition accuracy for A. viridis than CDA. In addition, use of 13 predictor variables yielded a greater identification accuracy than seven. The results of the current study showed that using seed morphological characteristics extracted by computer vision could be effective for reliable identification of the similar seeds of Amaranthus species.

Journal of Plant Protection Research

“…The neuron number of the hidden layer of the best network was determined, after constructing the networks. To find the most accurate networks in lentil yield prediction, different types of neural networks such as Multilayer Perceptrons, Generalized feedforward, Modular and Principal Component Analysis were trained and tested (Bagheri et al 2019). In addition, learning rules of Momentum, Levenberg Marquardt, Step, and Quickprop were tested.…”

“…In addition, learning rules of Momentum, Levenberg Marquardt, Step, and Quickprop were tested. The transfer functions of TanhAxon, SigmoidAxon, TanhAxon Linear, SigmoidAxon Linear, SoftMaxAxon, Linear Axon, and Axon also were evaluated (Bagheri et al 2019). Furthermore, the numbers of hidden layers and neurons in each hidden layer were manipulated to find the best neural networks.…”

“…Furthermore, the numbers of hidden layers and neurons in each hidden layer were manipulated to find the best neural networks. The above-mentioned steps to find the most accurate neural networks were done by trial and error (Niazian et al 2018;Bagheri et al 2019).…”