Abstract
Triplet loss function is one of the options that can significantly improve the accuracy of the One-shot Learning tasks. Starting from 2015, many projects use Siamese networks and this kind of loss for face recognition and object classification. In our research, we focused on two tasks related to vegetation. The first one is plant disease detection on 25 classes of five crops (grape, cotton, wheat, cucumbers, and corn). This task is motivated because harvest losses due to diseases is a serious problem for both large farming structures and rural families. The second task is the identification of moss species (5 classes). Mosses are natural bioaccumulators of pollutants; therefore, they are used in environmental monitoring programs. The identification of moss species is an important step in the sample preprocessing. In both tasks, we used self-collected image databases. We tried several deep learning architectures and approaches. Our Siamese network architecture with a triplet loss function and MobileNetV2 as a base network showed the most impressive results in both above-mentioned tasks. The average accuracy for plant disease detection amounted to over 97.8% and 97.6% for moss species classification.
Highlights
We tried several deep learning architectures and approaches
We believe that Siamese networks is a very promising direction, and we hoped that we could improve accuracy using the triplet loss function
When training the Siamese network with the triplet loss function, the input consists of three images, two of which belong to the same class, and the last one belongs to a different class
Summary
There are many studies in which DL models are used to detect and classify plant disease symptoms. Mohanty et al [7] use AlexNet and GoogLeNet on the PlantVillage database to classify 26 diseases and obtain 99.35 % accuracy on a test subset. The authors have not reported tests on real-life images; we can assume that the accuracy could be much worse Arguments for this assumption can be found in [9], where the author uses AlexNet, AlexNetOWTBn, GoogLeNet, Overfeat, VGG architectures on the second edition of the PlantVillage database with 87,848 images of 58 different classes. Türkoğlu and Hanbay [11] use various base networks such as AlexNet, VGG16, VGG19, SqueezeNet, GoogleNet, Inceptionv, InceptionResNetv, ResNet, ResNet101, and classifiers such as Knearest neighbor (KNN), Support vector machine (SVM), Extreme learning machine (ELM) Their self-collected database comprises 1965 high-resolution images of eight different plant diseases of four crops. We believe that Siamese networks is a very promising direction, and we hoped that we could improve accuracy using the triplet loss function
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
