DETECTION AND CLASSIFICATION OF PEST IN CROPS USING SINGLE SHOT MULTI-BOX DETECTOR.

Nowadays, according to the Food and Agriculture Organization (FAO), agricultural pests cause severe loss of global crop production every year. In order to eliminate these harmful insect pests, smart agriculture is the best way for farmers to use artificial intelligence techniques integrated with modern information and communication technology. Thus, they can increase the productivity of their crops. The main limiting factor for cultivation, however, is pest and disease infections. It is not possible to eradicate these diseases, but they can be controlled and monitored to minimize their impact. Crop diseases can be rapidly diagnosed by using automated image identification systems. An accurate identification system requires the extraction of features from images. To identify the samples, we used three different classes of machine learning: a support vector machine (SVM), a random forest (RF), and an artificial neural network (ANN). In addition, our deep learning method outperforms other previous recognition methods, namely Single Shot Multi-Box Detector (SSD), which increases the accuracy and multi-directional way of pest detection. Our results contribute to the use of machine learning in agriculture literature and have implications for research in deep learning and practical implications. [ABSTRACT FROM AUTHOR]