Deep Learning-Based Analysis of Crop Pathogen Using CNN

Deep Learning-Based Analysis of Crop Pathogen Using CNN

DINESH K

Department of Artificial Intelligence and Data Science
Panimalar Institute Of Technology Chennai, Tamil Nadu, India
milkydk10@gmail

Babisha A,
Assistant Professor
Department of Artificial Intelligence and Data Science
Panimalar Institute Of Technology Chennai, Tamil Nadu, India babisha15@gmail.com

HARIHARAN R

Department of Artificial Intelligence and Data Science
Panimalar Institute Of Technology Chennai, Tamil Nadu, India
haranhari1402@gmail.com

Dr. Suma Christal Mary S
Head of Department -IT
Department of Information Technology
Panimalar Institute Of Technology Chennai, Tamil Nadu, India ithod@pit.ac.in

MANI R

Department of Artificial Intelligence and Data Science
Panimalar Institute Of Technology Chennai, Tamil Nadu, India
maniix43@gmail.com

Swagatha J P,

Assistant Professor
Department of Information
Technology
Panimalar Institute Of Technology Chennai, Tamil Nadu, India swagathajaisathish@gmail.com

Abstract—This paper presents an advanced approach for detecting crop pathogens using deep learning techniques, specifically Convolutional Neural Networks (CNNs). Agricultural crop diseases pose a significant threat to global food security, neces- sitating rapid and accurate detection methods. Leveraging the power of deep learning, our proposed system utilizes enhanced CNN architectures to analyze images of diseased crops captured in the field. The CNNs are trained on large datasets of annotated images encompassing various crop species and pathogen types, enabling robust and generalized detection capabilities. Through extensive experimentation and evaluation, we demonstrate the effectiveness of our approach in accurately identifying crop pathogens with high precision and recall rates. Furthermore, we explore techniques to enhance model performance in sce- narios with limited training data and address challenges such as variability in image quality and complex visual similarities between different diseases. Our results indicate that CNN-based methods offer a promising solution for early and automated detection of crop diseases, facilitating timely interventions and ultimately contributing to sustainable agriculture practices and food security.

Keywords

Crop pathogens, Deep Learning (DL), Artificial Intelligence (AI), Convolutional Neural Networks (CNN), TensorFlow, Med- ical image analysis, Automated diagnosis, Transfer learning, Agricultural diagnostics.