Plant Disease Identification Based on Deep Learning Algorithm in Smart Farming

Yan Guo,Jin Zhang,Xiaonan Hu,Chengxin Yin,Wei Wang,Zhipeng Xue,Yu Zou

doi:10.1155/2020/2479172

Yan Guo, Jin Zhang + Show 5 more

Open Access

https://doi.org/10.1155/2020/2479172

Copy DOI

Abstract

The identification of plant disease is the premise of the prevention of plant disease efficiently and precisely in the complex environment. With the rapid development of the smart farming, the identification of plant disease becomes digitalized and data-driven, enabling advanced decision support, smart analyses, and planning. This paper proposes a mathematical model of plant disease detection and recognition based on deep learning, which improves accuracy, generality, and training efficiency. Firstly, the region proposal network (RPN) is utilized to recognize and localize the leaves in complex surroundings. Then, images segmented based on the results of RPN algorithm contain the feature of symptoms through Chan–Vese (CV) algorithm. Finally, the segmented leaves are input into the transfer learning model and trained by the dataset of diseased leaves under simple background. Furthermore, the model is examined with black rot, bacterial plaque, and rust diseases. The results show that the accuracy of the method is 83.57%, which is better than the traditional method, thus reducing the influence of disease on agricultural production and being favorable to sustainable development of agriculture. Therefore, the deep learning algorithm proposed in the paper is of great significance in intelligent agriculture, ecological protection, and agricultural production.

Highlights

Plant disease can directly lead to stunted growth causing bad effects on yields [1,2,3]
A novel plant leaf identification model based on deep learning algorithm is designed to solve the above issues. e function contains leaf retrieval, image segmentation, and identification with the utilization of integrated deep learning algorithm throughout the whole process. e first task is leaf retrieval, but many factors pose the challenge of identification accuracy such as soil and illumination in the complex environment [12]
3.1. e Solution Framework. e full plant disease identification model framework based on deep learning is shown in Figure 1, including three steps, the localization of plant leaves, the segmentation of images, the extraction of plant disease, and the identification of disease. e model used in this paper mainly consists of the following three steps. e first step is to locate the diseased leaves. e region proposal network (RPN) algorithm is used to train the leaf dataset in the complex environment, and the frame regression neural network and classification neural network are used to locate and retrieve the diseased leaves in the complex environment. e second step is the segmentation of diseased leaves. e Chan–Vese algorithm is used to segment the image of diseased leaves

Summary

Introduction

Plant disease can directly lead to stunted growth causing bad effects on yields [1,2,3]. A novel plant leaf identification model based on deep learning algorithm is designed to solve the above issues. E function contains leaf retrieval, image segmentation, and identification with the utilization of integrated deep learning algorithm throughout the whole process. E first task is leaf retrieval, but many factors pose the challenge of identification accuracy such as soil and illumination in the complex environment [12]. Image segmentation is the second step that is considered to be the most crucial because diagnostic precision plays an important role in detection results. E Chan–Vese algorithm based on region shows promising results for segmenting images free. E last step is to identify the disease of leaves based on the migration learning algorithm. Based on the pretrained model, the migration learning model uses the dataset of disease leaves in a simple background to train the model.

Literature Review

Modeling

Experimental Study

Findings

Results