A Nutrient Deficiency Prediction Method Using Deep Learning on Development of Tomato Fruits

During the development of tomato (Solanum lycopersicum L.) crop, the mineral nutrients are the important elements in the process of plant growth. Therefore, the recognition and prediction of the nutrient deficiencies in the cropping process have been great interest. In this paper, we propose a method of recognition and prediction the nutrient deficiency occurs on fruiting phrase of tomato plant based on the deep neural network. Moreover, we decide to use two essential mineral nutrients (i.e. Calcium and Potassium) for evaluating the nutrient status in the development of tomato plant. Inception-ResNet v2 based-Convolution Neural Network (CNN) is applied to distinguish the above mineral nutrients with the captured images of tomato plant growth under the greenhouse conditions. The aim of this study is to improve the accurate prediction of the nutrient deficiency for increasing the crop production and prevent the emergence of tomato pathology caused by the lack of nutrient. The performance of the Inception-ResNet v2 is validated through the real fruit images that were captured on growth of tomato plant.