Citation

Chia, Yu Thong and Priya Thiagarajah, Siva (2025) Segmentation-Based Detection of Oyster Mushroom Caps Using YOLOv8 for Smart Cultivation Monitoring. In: 2025 Multimedia University Engineering Conference, MECON 2025, 21 July 2025 - 23 July 2025, Cyberjaya, Malaysia.

Text 471.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Oyster mushroom (Pleurotus Ostreatus) cultivation is gaining popularity due to its short growth cycle, rich nutrient contents and easy growing. This research proposes a computer vision-based methodology where the YOLOv8 segmentation model has been used to identify oyster mushroom caps from their cultivation bed image data. A manual-annotated polygon-based annotation dataset has been used to train the YOLOv8 segmentation model. This model also provides infestation detection which enables the system to have targeted treatment at an early stage to minimize overall damage and maintain the productivity of the cultivation environment. The problem to be addressed through this study is to determine the optimal harvesting time for oyster mushrooms. This is to prevent overgrown mushroom, contamination and infestation that lead to poor quality of mushroom production. Suitable design of computer vision model is the proposed solution to provide a real-time visual tracking pipeline for oyster mushroom growth monitoring. The trained model achieves mean Average Precision (mAP@0.5) of 91.9% and (mAP@0.5–0.95) of 79.1%. The model’s ability to mask a single mushroom cap size area was established using test images of different classes, from the beginning stage of small-cap size to ready harvest till overgrown large mushroom cap and that resulted in detection. This work explains the setup of the model pipeline which consists of data preparation, model training, and inference. Early post-processing was explored, and the trained model is set out to build a reliable detection framework to be used as a reference point for future maturity analysis, tracking, and harvesting reasoning. The results demonstrate the potential of using deep learning segmentation in mushroom cultivation to enable scalable low-labour monitoring.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	YOLOv8, Instance Segmentation, Mushroom Monitoring, Smart Farming, Computer Vision, Precision Agriculture.
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD1401-2210 Agriculture
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	19 Mar 2026 00:54
Last Modified:	19 Mar 2026 00:54
URII:	http://shdl.mmu.edu.my/id/eprint/15499

Downloads

Edit (login required)