Detection of Cucumber Fruits with Excessive Consumption of Nitrogen using Hyperspectral imaging (With Emphasis on Sustainable Agriculture)

Uncontrolled consumption of nitrogen in cucumber fruit causes nitrate accumulation in the fruit that is harmful for humans and ecosystem. One of precise way to diagnose nutritional disorders is plant analysis, which is a destructive, costly, and time-consuming method. Hence, present study aims to evaluate cucumbers non-destructively using hyperspectral imaging. Cucumber seeds were planted in 16 pots and after the growth of plants and the appearance of fruits, the pots were divided into 4 categories. One was considered as a control sample (with normal nitrogen) and rest were treated with excess nitrogen by 30%, 60% and 90%, respectively. Hyperspectral images were obtained during two stages namely before treatment and 10 days after treatment. Three proposed methods namely hybrid neural network-cultural (ANN-CA) algorithm, multilayer perceptron neural network (MLP) and support vector machine (SVM) were used to analyse and classify fruits. The findings represented the correct classification rates of 92%, 89.51% and 78.97% for ANN-CA, MLP and SVM, respectively. Thus, the ANN-CA algorithm had a good ability to identify excess nitrogen fruits.