Identification of Tomato Disease Types and Detection of Infected Areas Based on Deep Convolutional Neural Networks and Object Detection Techniques.

Qimei Wang,Minghe Sun,Feng Qi,Jianhua Qu,Jie Xue

doi:10.1155/2019/9142753

Abstract

This study develops tomato disease detection methods based on deep convolutional neural networks and object detection models. Two different models, Faster R-CNN and Mask R-CNN, are used in these methods, where Faster R-CNN is used to identify the types of tomato diseases and Mask R-CNN is used to detect and segment the locations and shapes of the infected areas. To select the model that best fits the tomato disease detection task, four different deep convolutional neural networks are combined with the two object detection models. Data are collected from the Internet and the dataset is divided into a training set, a validation set, and a test set used in the experiments. The experimental results show that the proposed models can accurately and quickly identify the eleven tomato disease types and segment the locations and shapes of the infected areas.

Highlights

Plant diseases have always been a thorny problem in agricultural production and one of the main factors restricting the sustainable development of agriculture
(3) Consult experts to undertake an analysis of the disease symptoms erefore, it is possible for a person with strong professional knowledge to accurately diagnose plant diseases
Object detection models and different deep convolutional neural networks (DCNNs) [1] are combined, which can not Computational Intelligence and Neuroscience only identify the types of tomato diseases, and locate the diseased spots so as to use the appropriate treatment

Summary

Introduction

Plant diseases have always been a thorny problem in agricultural production and one of the main factors restricting the sustainable development of agriculture. Object detection models and different deep convolutional neural networks (DCNNs) [1] are combined, which can not Computational Intelligence and Neuroscience only identify the types of tomato diseases, and locate the diseased spots so as to use the appropriate treatment. I.e., Faster R-CNN [2] and Mask R-CNN [3], are combined with four different deep convolutional neural networks. Deep convolutional neural networks are used to automatically extract original image features, and object detection architectures are used to identify, classify, and locate diseased sites in feature maps. E purpose of Faster R-CNN is to identify and locate diseased tomatoes, while that of Mask R-CNN is to segment specific lesion areas on diseased tomatoes (see Figure 1). Python scripts are used to mark the object and visually show the different purposes of the two architectures

Related Works

Results

Dataset

Qualitative Analysis

Discussion