ApaltAI: a web-based diagnostic system with a sequential voting architecture for detecting anthracnose and scab in...

Area of Science:

• Agricultural Science
• Computer Science
• Plant Pathology

Background:

• Avocado production faces significant economic losses (20-30%) due to diseases like anthracnose and scab.
• Accurate and timely disease identification is crucial for maintaining fruit quality and yield in avocado cultivation.

Purpose of the Study:

• To develop an automated system for identifying two major avocado diseases: anthracnose and scab.
• To integrate a binary sequential voting architecture (VotingBS) with a web application for practical disease diagnosis.

Main Methods:

• Implementation of a hierarchical two-stage deep learning ensemble (VotingBS) for disease classification.
• Training and validation using a dataset of 674 labeled avocado fruit images.
• Development of a web application with modules for crop management and disease diagnosis.

Main Results:

• The VotingBS system achieved high performance metrics: 98.92% precision, 98.89% recall, and 99.03% accuracy.
• The developed system significantly outperformed traditional disease identification methods.
• The integrated web application enhances practical utility for farmers and agricultural agencies.

Conclusions:

• The hybrid deep learning model combined with a digital platform offers a revolutionary approach to plant disease diagnostics.
• The system promotes efficient, automated, and resilient precision agriculture practices.
• This technology can significantly reduce economic losses in avocado production.