Plant Disease Detection On Some Selected Plants Using Convolutional Neural Network (CNN)

Hashim Bisallah Ibrahim & Ogbonna Onyedikachi Benjamin

Plant Disease Detection On Some Selected Plants Using Convolutional Neural Network (CNN)

Abstract: Crop diseases pose a severe danger to food security, yet because many parts of sub-Saharan Africa lack basic infrastructure, it is still challenging to identify them quickly. Smartphone-assisted illness recognition is now possible thanks to the growing popularity of smartphones and recent developments in computer vision made possible by deep learning. Using a locally collected dataset of 268 photos of both healthy and sick plant leaves of maize, beans, rice, guinea corn, and sunflower, aggregated at optimum environments, we trained a profound CNN to recognize plant illnesses. The developed model's response rate was 41.02%–94.03%, signifying the approach's usefulness. Generally, the practice of utilizing ImageNet open source data in profound models training, and learning avails a precise chronology for mobile devices embedded in plant illnesses identification and control in Nigeria, and the whole world.

JASIC Volume. 4, Issue 2 (2023)

Contributor(s)

Keywords

Download Full-text (PDF)

Plant Disease Detection On Some Selected Plants Using Convolutional Neural Network (CNN)