.

ISSN 2063-5346
For urgent queries please contact : +918130348310

RICE LEAF DISEASE RECOGNITION USING CNN

Main Article Content

Dr. V. Anantha Krishna, A. Shreya Reddy, Ch.Akshaya, K. Bhanusri
» doi: 10.31838/ecb/2023.12.s3.274

Abstract

Numerous bacterial, viral, or fungal conditions affect the rice splint, and they drastically lower rice product. The identification of rice leaf conditions is essential to meeting the demand for rice from a sizable worldwide population. Rice leaf disease can only be linked grounded on the backgrounds and circumstances of the image accession. By lowering the network parameters, we present a new CNN- grounded model to identify ails of rice leaves. Several CNN-based models are trained to recognize typical rice splint conditions using a fresh dataset of prints of rice splint conditions. The suggested model obtains the stylish confirmation delicacy of97.35 percent and training accuracy of 99.7 percent. These issues prove our strategy's energy and supremacy. With a topmost accuracy of 97.82 and an area under wind( AUC) of0.99, the effectiveness of the suggested model is assessed using a collection of independent rice leaf disorder. photos. Also, studies with this have been done. These outcomes show how our strategy outperforms cutting-edge CNN-based models for identifying rice leaf diseases, proving its usefulness.

Article Details