Analysis of Image Recognition Algorithms for Detecting Common Calamansi [Citrofortunella microcarpa (Bunge) Wijnands] Diseases in the Philippines

Rean T. Goloy; Concepcion L. Khan

Rean T. Goloy Mindoro State University, Alcate, Victoria, Oriental Mindoro, Philippines
Concepcion L. Khan University of the Philippines- Los Baños, Los Baños, Laguna, Philippines

Abstract

Purpose – Plant diseases are a major challenge for agriculture in the Philippines, particularly for calamansi [Citrofortunella microcarpa (Bunge) Wijnands], a key citrus crop. This study addresses this issue by employing image processing and machine learning techniques to enhance the detection of calamansi diseases.

Method – The research analyzed three image recognition algorithms—YOLOv5, Faster R-CNN, and MobileNetSSDv2—using a dataset of 2,990 images of calamansi leaves and fruits collected from farms in Victoria, Oriental Mindoro. The images were processed, and the models were evaluated based on precision, recall, and mean average precision (mAP).

Results – YOLOv5 outperformed the other models with a precision of 96%, recall of 96.2%, and mAP of 98.1%. This model was deployed in a mobile application, achieving a field-testing accuracy of 77.5%. The findings highlight the potential of using machine learning for real-time detection of plant diseases.

Conclusion—The study demonstrates the feasibility of integrating advanced algorithms into mobile technology to assist Calamansi farmers in monitoring crop health effectively.

Practical Implications – Adopting mobile-based disease detection systems can significantly improve crop management, productivity, and sustainability. However, the research is limited to specific diseases and data from a single region, underscoring the need for broader datasets and extended testing.

Author Biographies

Rean T. Goloy, Mindoro State University, Alcate, Victoria, Oriental Mindoro, Philippines

Rean Tabernero Goloy is passionate about advancing knowledge in artificial intelligence, machine learning, and digital image processing, with a focus on agricultural applications. He earned his B.S. in Information Technology magna cum laude from Mindoro State University (formerly Mindoro State College of Agriculture and Technology) in 2017, followed by a Master's in Computer Science from the University of the Philippines – Los Baños. Hailing from Naujan, Oriental Mindoro, he is the youngest of six children of Mr. Fernando Madriaga Goloy and Mrs. Marcelina Tabernero Goloy. His current research reflects his commitment to leveraging technology for agricultural innovation.

Concepcion L. Khan, University of the Philippines- Los Baños, Los Baños, Laguna, Philippines

Concepcion L. Khan is an Associate Professor based at the University of the Philippines, Los Baños. With a passion for advancing technology, her research focuses on Digital Agriculture, Data Mining, Artificial Intelligence, and Information Systems. She is dedicated to exploring innovative solutions to real-world challenges through data-driven insights and intelligent systems. Prof. Concepcion actively collaborates with students and colleagues to push the boundaries of research in her fields of expertise. She can be reached at clkhan@up.edu.ph for academic inquiries and collaborative opportunities.