Feasibility of Hybrid PSO-ANN Model for Identifying Soybean Diseases

Soybean disease has become one of vital factors restricting the sustainable development high-yield and high-quality soybean industry. A hybrid artificial neural network (ANN) model optimized via particle swarm optimization (PSO) algorithm, which is denoted as PSO-ANN, proposed in this paper for diseases identification based on categorical feature inputs. Augmentation dataset created Synthetic minority over-sampling technique (SMOTE) to deal with quantitative insufficiency unbalance dataset. PSO algorithm used optimize parameters ANN, including activation function, number hidden layers, neurons each layer optimizer. In end, ANN 2 63 61 layers respectively, Relu function Adam optimizer yields best overall test accuracy 92.00%, compared traditional machine learning methods. PSO-ANN shows superiority various evaluation metrics, may have great potential crop control modern agriculture.