International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 05 | May 2023
p-ISSN: 2395-0072
www.irjet.net
Potato leaf disease detection using convolutional neural networks P Radhika1, P Tejeswara Murthy 2, G Pavaneeshwar Reddy3, V Durga Vara Prasad4, T Harshitha5, N B V Bharath6 1 Associate Professor, Department of Computer Science and Engineering 2,3,4,5,6 Undergraduate Student, Department of Computer Science and Engineering, Vallurupalli
Nageswara Rao Vignana Jyothi Institute of Engineering and Technology, Hyderabad, Telangana, India ---------------------------------------------------------------------***--------------------------------------------------------------------The goal of this study is to train a CNN model on a Abstract - Crop diseases significantly damage agriculture,
collection of photos of potato leaves in order to identify and categorize three different potato leaf diseases: lateblight, earlyblight, and healthy. Multiple layers make up the model's architecture, including convolutional and pooling layers that take relevant characteristics from the input images and process them. The model also includes a data augmentation process to boost its performance even more.
hurting livelihoods and the stability of the economy. In this study, a unique method for identifying the three potato plant diseases Lateblight, Earlyblight, and Healthy was established using Convolutional Neural Networks (CNNs). The model's ability to identify diseases showed outstanding accuracy. Assessments of the viability found easy integration into current systems at reasonable implementation costs. Stakeholders gave the model positive feedback and acknowledged its usefulness in making decisions. This study highlights how deep learning models have the ability to successfully manage illnesses, reduce crop losses, and enhance overall agricultural health.
The trained model is assessed for its efficiency in identifying and categorizing potato leaf diseases using a variety of performance indicators, such as accuracy, precision, recall, and F1-score. Additionally, the model's performance on unknown data is assessed using realworld validation datasets, which sheds light on its generalization potential and practical application.
Key Words: Neural Network, Artificial Intelligence (AI) Classification, Deep Learning, Image Augmentation, CNN, GDP.
The results of this study are anticipated to aid in the creation of effective and automated systems for controlling potato leaf disease. Deep learning models can accurately and quickly detect diseases, allowing farmers and other agricultural stakeholders to implement targeted interventions like disease-specific treatments and preventive measures, improving crop health, yields, and agricultural sustainability.
1. INTRODUCTION Potato plants are important food crops that play a vital role in both global food security and economic expansion. However, these crops are prone to a number of illnesses, which can significantly impair crop quality and output. For the purpose of implementing focused disease management methods and minimizing the detrimental effects on crop output, prompt and accurate diagnosis of potato leaf diseases is essential.
In conclusion, this research uses data augmentation and CNN-based deep learning approaches to create a reliable model for detecting potato leaf disease. The results of this study could significantly alter how potato crops are managed in the future, reduce production losses, and encourage the use of sustainable farming methods.
Convolutional Neural Networks (CNNs), one of the most recent developments in deep learning techniques, have completely changed the way that object detection and picture classification are done. These methods have demonstrated a lot of promise for correctly recognizing and categorizing intricate patterns and features in photos. Researchers have created a robust CNN-based model for detecting potato leaf disease by utilizing the power of deep learning.
2. RELATED WORK A deep learning-based plant disease detection and diagnostic system for crop protection was suggested in the paper by Zhang et al. (2020). Convolutional Neural Networks (CNNs), among other deep learning approaches, were used by the researchers to preprocess photos, extract features, and categorize plant illnesses. The created model successfully detected a variety of plant diseases with an excellent accuracy rate of 95%. Their research demonstrated how deep learning models can accurately identify and diagnose plant diseases, which can
To improve the model's ability to generalize and identify diseases under varied circumstances, the suggested model utilizes data augmentation techniques. In order to increase the diversity of the training data and increase the resilience of the model, data augmentation entails adding random transformations to the input images, such as flips, rotations, zooms, and rescaling.
© 2023, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 1191