Skip to main content
ARS Home » Southeast Area » Auburn, Alabama » Soil Dynamics Research » Research » Publications at this Location » Publication #430572
Research Project: Sustaining Productivity and Ecosystem Services of Agricultural and Horticultural Systems in the Southeastern United States

Location: Soil Dynamics Research

Title: MagCount: A digital framework for automated counting of southern magnolias for inventory management in nurseries

Author
item SYED, M - The University Of Texas At Dallas
item MANJUNATHA, H - University Of Texas
item BORAH, S - University Of Texas
item SUNDARAVADIVEL, P - University Of Texas
item KNIGHT, P - Mississippi State University
item Kumpatla, Siva Prasad
item GOWDA, P - University Of Texas At Arlington
item Torbert Iii, Henry

Submitted to: Meeting Abstract
Publication Type: Proceedings
Publication Acceptance Date: 12/20/2025
Publication Date: N/A
Citation: N/A

Interpretive Summary: For the green industry, accurate inventory tracking is essential. However, manual plant counts can be both costly and prone to errors. MagCount is an innovative framework that employs a fully automated approach to inventory management through a combination of aerial robot imagery and artificial intelligence. A study on southern magnolia (Magnolia grandiflora) using a machine vision algorithm called YOLOv8 attained a 95.2% accuracy in counting southern magnolia specimens compared to only 41.7% accuracy using conventional YOLO algorithm. The model was trained with 561 high-resolution images. This system combines object recognition with machine learning, enabling image tiling, which offers advantages for commercial nurseries by reducing operational cost as well as minimizing human error.

Technical Abstract: For the green industry, accurate inventory tracking is essential. However, manual plant counts can be both costly and prone to errors. MagCount is an innovative framework that employs a fully automated approach to inventory management through a combination of aerial robot imagery and artificial intelligence. A study on southern magnolia (Magnolia grandiflora) using a machine vision algorithm called YOLOv8 attained a 95.2% accuracy in counting southern magnolia specimens compared to only 41.7% accuracy using conventional YOLO algorithm. The model was trained with 561 high-resolution images. This system combines object recognition with machine learning, enabling image tiling, which offers advantages for commercial nurseries by reducing operational cost as well as minimizing human error.