DIAGNOSIS OF PLANT DISEASE USING DEEP LEARNING

• Published in: I-manager's Journal on Pattern Recognition Vol. 8; no. 1; p. 25
• Main Authors: SHANMUGARAJ, G., RAMYA, G., SOWMIYA, D., VINOTHINI, S.
• Format: Journal Article
• Language: English
• Published: Nagercoil iManager Publications 2021
• Subjects: Artificial neural networks; Crop diseases; Datasets; Deep learning; Neural networks; Pesticides; Plant diseases
• ISSN: 2349-7912; 2350-112X
• DOI: 10.26634/jpr.8.1.18228

Crop plants play a vital role in the field of agriculture and also influence climatic change, and therefore, taking care of crop plants is very essential. Similar to humans, plants are also affected by many diseases caused by bacteria, fungi and virus. Recognizing of these diseases at the right time and restoring them is very essential to prevent whole crop from destruction. This paper suggests a deep learning-based model plant to detect plant disease in crops. The model would use neural network and would be able to identify several diseases from plants using images of their leaves. Augmentation is applied on dataset to increase the sample size. Then Convolution Neural Network (CNN) is used with multiple pooling and convolution layers. Plant dataset is used to instruct the model. After training the model, it is tested to validate the output. Based on the trained dataset, we classify the disease according to the class to which it belongs. In our future work we would consider suggesting appropriate pesticides provided to eradicate the identified disease. Also, in future our model can be integrated with drone to detect plant diseases.