A Comparative Study of Deep CNN in Forecasting and Classifying the Macronutrient Deficiencies on Development of Tomato Plant

Abstract

During the process of plant growth, such as during the flowering stages and fruit development, the plants need to be provided with the various minerals and nutrients to grow. Nutrient deficiency is the cause of serious diseases in plant growth, affecting crop yield. In this article, we employed artificial neural network models to recognize, classify, and predict the nutritional deficiencies occurring in tomato plants (Solanum lycopersicum L.). To classify and predict the different macronutrient deficiencies in the cropping process, this paper handles the captured images of the macronutrient deficiency. This deficiency during the fruiting and leafing phases of tomato plant are based on a deep convolutional neural network (CNN). A total of 571 images were captured with tomato leaves and fruits containing the crop species at the growth stage. Among all images, 80% (461 captured images) were used for the training dataset and 20% (110 captured images) were applied for the validation dataset. In this study, we provide an analysis of two different model architectures based on convolutional neural network for classifying and predicting the nutrient deficiency symptoms. For instance, Inception-ResNet v2 and Autoencoder with the captured images of tomato plant growth under the greenhouse conditions. Moreover, a major type of statistical structure, namely Ensemble Averaging, was applied with two aforementioned predictive models to increase the accuracy of predictive validation. Three mineral nutrients, i.e., Calcium/Ca2+, Potassium/K+, and Nitrogen/N, are considered for use in evaluating the nutrient status in the development of tomato plant with these models. The aim of this study is to predict the nutrient deficiency accurately in order to increase crop production and prevent the emergence of tomato pathology caused by lack of nutrients. The predictive performance of the three models in this paper are validated, with the accuracy rates of 87.273% and 79.091% for Inception-ResNet v2 and Autoencoder, respectively, and with 91% validity using Ensemble Averaging.

Deep Learning for Non-Invasive Diagnosis of Nutrient Deficiencies in Sugar Beet Using RGB Images

“…Note that our datasets are similar in terms of number of images and treatments, but their reported accuracy is not promising, possibly due to the fact that the dataset they used only contains images of leaves, where pathological symptoms are easily misinterpreted. Tran et al [ 29 ] classified three nutrient deficiencies for 571 tomato images using an ensemble of an Inception-ResNet and an autoencoder, achieving an accuracy of 91% in the test set. However, the dataset that they used is small-scale and only contains treatments of N, K, Ca.…”

Classifying the Macronutrient Deficiency in Soybean Leaf with Deep Learning

“…Quatro trabalhos foram relacionados [Wulandhari et al 2019, Tran et al 2019, Watchareeruetai et al 2018, Hasan et al 2018] com relação no uso de RNC, imagens e algum tipo de classificação com base em macronutrientes. Em [Tran et al 2019], apresentam uma pesquisa para classificar e prever diferentes deficiências de macronutrientes (cálcio, potássio e nitrogênio) no processo de cultivo do tomate. Os autores também fizeram o uso de data augmentation para aumentar o conjunto de imagens.…”

“…Os resultados são apresentados apenas com três métricas, devido a inexistência da especificidade nos demais trabalhos. Vale ressaltar que o trabalho dessa pesquisa analisa um macronutriente, o potássio, os trabalhos [Wulandhari et al 2019, Tran et al 2019, Watchareeruetai et al 2018 analisam mais de um macronutriente. Já em [Hasan et al 2018] não entra em detalhes sobre qual macronutriente foi analisado nos resultados.…”

“…A grande quantidade de imagens também teve grande relação na precisão do método, em diversos trabalhos [Wulandhari et al 2019, Dosovitskiy et al 2014] mostram que o aumento do dataset contribui para a precisão da RNC, e a augmentationé um bom caminho caso o dataset seja pequeno. Por exemplo, em [Tran et al 2019] poucas unidades classificadas de forma errada têm grande impacto na precisão e acurácia da proposta. Com o aumento da quantidade de imagens, a capacidade de generalização da RNCé maior e consequentemente melhora o seu desempenho.…”

EEkNN: k-Nearest Neighbor Classifier with an Evidential Editing Procedure for Training Samples

“…Classification of patterns is an important area of research and practical applications in a variety of fields including biology [1], psychology [2], medicine [3], electronics [4], marketing [5], military affairs [6], etc. In the past several decades, a wide variety of approaches has been developed towards this task [7].…”