Result: CFUs detection in Petri Dish images using YOLOv12

Title:
CFUs detection in Petri Dish images using YOLOv12
Authors:
Quevit, Victorien, Dillenseger, J-L, Laferté, J-M, Fougères, A-J, Djelal, H, Jalenques, E
Contributors:
Dillenseger, Jean-Louis, IFSA Publishing
Publisher Information:
2025.
Publication Year:
2025
Subject Terms:
Petri Dish, [INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], Deep Learning, [INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], YOLO, Convolutional Neural Network, [SDV.MP] Life Sciences [q-bio]/Microbiology and Parasitology, Colony Forming Units, [SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing
Document Type:
Conference Conference object
File Description:
application/pdf
Language:
English
Access URL:
https://hal.science/hal-05024179v1
Accession Number:
edsair.od......2755..1e009ffdc890b10ba291b12feedc593c
Database:
OpenAIRE

Further Information

This study proposes to use the Deep Learning detection model, YOLOv12, for Colony-Forming Unit (CFU) detection in Petri dish images, aiming to automate the traditionally labor-intensive and error-prone manual counting process. YOLOv12 integrates attention mechanisms to enhance detection accuracy while maintaining real-time performance. The model achieves a mAP50 of 0.975 and a mAP50:95 of 0.706 across all 5 classes of CFU in the AGAR dataset, demonstrating its effectiveness in automating microbiological analysis. This innovation highlights the potential of YOLOv12 to streamline laboratory workflows and improve accuracy in CFU detection.