Hybrid CNN & Random Forest Model for Effective Carom Leaf Disease Diagnosis

This study presents an advanced machine learning system for accurately classifying illnesses in carom leaves. The model displays amazing success in distinguishing between distinct carom leaf illnesses by using a wide collection of parameters such as precision, recall, F1-score, assistance, or accura...

Full description

Saved in:
Bibliographic Details
Main Authors: Banerjee, Deepak, Sharma, Neha, Chauhan, Rahul, Singh, Mukesh, Gill, Kanwarpartap Singh
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
Crops
Diseases
Forest
Medical diagnosis
Microorganisms
Numerical models
Pathology
Precision
Sensitivity
Solid modeling
Solids
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study presents an advanced machine learning system for accurately classifying illnesses in carom leaves. The model displays amazing success in distinguishing between distinct carom leaf illnesses by using a wide collection of parameters such as precision, recall, F1-score, assistance, or accuracy. Powdery Mildew, Leaf Spot Diseases, Root Rot, Downy Mildew, Viral Infections, Aphid Infestation, Bacterial Blight, and Healthy Crops are all covered in the research. The numerical values shown highlight the model's resilience, with every illness category demonstrating high precision, recall, or F1-score percentages. The support metric offers details about the data set's distribution, with an overall accuracy of 99%. The model's capacity to retain consistency in disease diagnosis is highlighted by the macro and average weighted metrics, which show a balanced outcome across all classes. The micro average statistic confirms the model's overall performance in the categorization task. This study makes a substantial contribution to the science of plant pathology by providing an accurate and precise method for recognizing and treating illnesses in carom leaves using advanced artificial intelligence techniques.
ISSN:2469-5556
DOI:10.1109/ICACCS60874.2024.10717295