Abstract

Plants rely on a delicate balance of 16 essential nutrients to thrive, with macronutrients being crucial for robust growth, while micronutrients play a vital role despite being needed in smaller quantities. Insufficient nutrient levels can stunt plant growth, hinder flowering, and reduce fruit yield. Accurate diagnosis of these deficiencies is paramount for farmers to address issues effectively, ensuring the cultivation of nutrient-rich crops and maximizing yield. Bananas, a globally significant fruit crop known for its high nutritional value, require meticulous nutrient management to thrive. Micronutrients like Boron, are critical for maintaining hormonal equilibrium in banana plants, with deficiencies often manifesting visibly on leaves. This study proposes a deep-learning approach to detect Boron deficiencies in banana leaves. The developed CNN model with Skip Connections (CNNSC), comprising thirteen layers, outperforms established architectures like VGG16, DenseNet, and Inception V3. Evaluation metrics showcase the model’s effectiveness, achieving a remarkable accuracy of approximately 95%.

Details

Title
Boron Deficiency Detection in Banana Leaves using Skip-Connected Convolutional Neural Network (SC-CNN)
Author
Sunitha, P; Geetha, Kiran A; Uma; Channakeshava; Babu, Suresh
Pages
1-24
Section
Research Articles
Publication year
2024
Publication date
2024
Publisher
CHRIST (Deemed to be University)
e-ISSN
09753303
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
3205315793
Copyright
© 2024. Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the associated terms available at http://journals.christuniversity.in/index.php/mapana/about