Deep convolutional neural networks for mobile capture device-based crop disease classification in the wild

Picón, Artzai; Alvarez-Gila, Aitor; Seitz, Maximiliam; Ortiz‐Barredo, Amaia; Echazarra, Jone; Johannes, Alexander

doi:10.1016/j.compag.2018.04.002

Cited by 344 publications

(171 citation statements)

References 42 publications

Supporting

Mentioning

147

Contrasting

Unclassified

Order By: Relevance

“…It might be more useful to enhance the dataset size. 23 In an unusual strategy, the adapted Deep Residual Neural Network (CNN)-based algorithm, 24 namely ResNet50, 28 was integrated for mobile application for early identification of three European endemic wheat diseases: Septoria, Tan Spot, and Rust.…”

Section: Discussion and Comparisonmentioning

confidence: 99%

“…It might be more useful to enhance the dataset size . In an unusual strategy, the adapted Deep Residual Neural Network (CNN)‐based algorithm, namely ResNet50, was integrated for mobile application for early identification of three European endemic wheat diseases: Septoria, Tan Spot, and Rust. In further experiments on Septoria, Tan Spot, Rust, and healthy types, the use of ResNet50 achieved a balanced accuracy rate of 84.00%; this score improved to a 96.00% detection rate with the fully convolutional network (FCN) for disease detection in a pilot region of Germany.…”

Section: System Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

Identification of wheat kernels by fusion of RGB, SWIR, and VNIR samples

2019

View full text Add to dashboard Cite

Background The sustainable management of agricultural resources requires the integration of cutting‐edge science with the observation and identification of crops. This assists experts to make correct decisions. The aim of this study is to assess the robustness of a commonly used deep learning tool, VGG16, in improving the categorization of wheat kernels. Two fusion methodologies were considered simultaneously. We performed experiments on visible light (RGB), short wave infrared (SWIR), and visible‐near infrared (VNIR) datasets, including 40 classes, with 200 samples in each class, giving 8000 samples in total. Results After making simulations with 6400 training and 1600 testing samples, we achieved excellent performance scores, with 98.19% and 100% accuracy rates, respectively. Conclusion The wheat identification system developed here serves as an effective identification framework and supports the view that deep learning tools can adequately discriminate between different types of wheat kernels. The proposed automated system would be useful for improving economic growth and in reducing the labor force, leading to greater efficiency and higher productivity in the wheat industry. © 2019 Society of Chemical Industry

show abstract

Section: Discussion and Comparisonmentioning

confidence: 99%

Section: System Evaluationmentioning

confidence: 99%

Identification of wheat kernels by fusion of RGB, SWIR, and VNIR samples

2019

View full text Add to dashboard Cite

show abstract

“…If nothing but expert annotations will suffice, data sharing lessens the burden on any one group. Multiple groups have used the International Skin Imaging Collaboration image set of human skin diseases (Codella et al, 2015;Haenssle et al, 2018) or the PlantVillage image set of plant diseases (Mohanty et al, 2016;Ghosal et al, 2018;Picon et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Millimeter-Level Plant Disease Detection From Aerial Photographs via Deep Learning and Crowdsourced Data

Wiesner-Hanks

Stewart

et al. 2019

Front. Plant Sci.

View full text Add to dashboard Cite

Computer vision models that can recognize plant diseases in the field would be valuable tools for disease management and resistance breeding. Generating enough data to train these models is difficult, however, since only trained experts can accurately identify symptoms. In this study, we describe and implement a two-step method for generating a large amount of high-quality training data with minimal expert input. First, experts located symptoms of northern leaf blight (NLB) in field images taken by unmanned aerial vehicles (UAVs), annotating them quickly at low resolution. Second, non-experts were asked to draw polygons around the identified diseased areas, producing high-resolution ground truths that were automatically screened based on agreement between multiple workers. We then used these crowdsourced data to train a convolutional neural network (CNN), feeding the output into a conditional random field (CRF) to segment images into lesion and non-lesion regions with accuracy of 0.9979 and F1 score of 0.7153. The CNN trained on crowdsourced data showed greatly improved spatial resolution compared to one trained on expert-generated data, despite using only one fifth as many expert annotations. The final model was able to accurately delineate lesions down to the millimeter level from UAV-collected images, the finest scale of aerial plant disease detection achieved to date. The two-step approach to generating training data is a promising method to streamline deep learning approaches for plant disease detection, and for complex plant phenotyping tasks in general.

show abstract

“…Forward Chaining [1], Fuzzy system [2], K-Means Clustering [3] [4], Decision Tree [5] [6], Bayesian networks and incremental learning [7], Naïve Bayes and Certainty Factor [8], Naïve Bayes [9], Computer vision and artificial intelligence [10], Deep Convolutional Neural Network [11] [12], Fuzzy inference system [13], Convolutional Neural Network (CNN) [14], Fractal Dimension Values and Fuzzy C-Means [15], Deep learning [16] [17], Machine learning [18]. Furthermore, there are some applications for detection system of plant pests and diseases using technologies as follows: expert system [1] [19], mobile system [6] [12], computer vision and artificial intelligence [10] [20], image processing system [15] [16] [18] [20], and Internet of Things (IoT) [21] [22] [23]. However, research in this area is still needed, especially from the computer science perspective.…”

Section: Introductionmentioning

confidence: 99%

Technologies, methods, and approaches on detection system of plant pests and diseases

Anamisa¹,

Yusuf²,

Agustiono³

et al. 2019

EECSI

View full text Add to dashboard Cite

This research aims to identify the technology, methods, approaches applied in developing plant pest and disease detection systems. For this purpose, it mainly reviews systematically related research on identification, monitoring, detection, and control techniques of plant pests and diseases using a computer or mobile technology. Evidence from the literature shows previous both academia and practitioners have used various technologies, methods and approaches for developing detection system of plant pests and diseases. Some technologies have been applied for the detection system, such as web-based, mobile-based, and internet of things (IoT). Furthermore, the dominant approaches are expert system and deep learning. While backward chaining, forward chaining, fuzzy model, genetic algorithm (GA), K-means clustering, Bayesian networks and incremental learning, Naïve Bayes and Certainty Factors, Convolutional Neural Network, and Decision Tree are the most frequently methods applied in the previous researches. The review also indicated that no single technology or technique is best for developing accurate pest/disease detection system. Instead, the combination of technologies, methods, and approaches resulted in different performance and accuracies. A possible explanation for this is because the systems are used for detecting, controlling and monitoring various plants, such as corn, onion, wheat, rice, mango, flower, and others that are different. This research contributes by providing a reference for technologies, methods, and approaches to the detection system for plant pests and diseases. Also, it adds a way of literature review. This research has implications for researchers as a reference for researching in the computer system, especially for the detection of plant pest and disease research. Hence, this research also extends the body of knowledge of the intelligence system, deep learning, and computer science. For practice, the method references can be used for developing technology for detecting plant pest and disease.

show abstract

Deep convolutional neural networks for mobile capture device-based crop disease classification in the wild

Cited by 344 publications

References 42 publications

Identification of wheat kernels by fusion of RGB, SWIR, and VNIR samples

Identification of wheat kernels by fusion of RGB, SWIR, and VNIR samples

Millimeter-Level Plant Disease Detection From Aerial Photographs via Deep Learning and Crowdsourced Data

Technologies, methods, and approaches on detection system of plant pests and diseases

Contact Info

Product

Resources

About