I+R+D

Computer vision to identify marine microalgae in water samples

Photo of author

By Milthon Lujan

Microscopic imaging combined with computer vision and microfluidic technology is used for real-time detection and health monitor of algae. Source: Zhou et al., (2023)
Microscopic imaging combined with computer vision and microfluidic technology is used for real-time detection and health monitor of algae. Source: Zhou et al., (2023)

Despite scientific and technological advances, the main method of identifying and counting algae species is primarily through observation under a microscope, which is time-consuming and laborious.

To monitor and determine the health status of marine microalgae in situ, hardware and software integration is needed to enable real-time algal imaging, data analysis, and implementation of machine learning algorithms to recognize and classify the cells of marine microalgae.

A team of researchers from Hainan University, Zhejiang University and The State University of New York at Stony Brook proposes a new database for the detection of marine microalgae and a set of detection methods.

The database includes images of six genera of microalgae (Pinnularia, Chlorella, Platymonas, Dunaliella salina, Isochrysis, and Symbiodinium) that are common in the ocean, and of the different developmental stages of the genera of microalgae.

Computer vision to identify microalgae

Object detection is a computer vision task that involves the location and identification of objects in an image or video.

The researchers applied computer vision algorithms for the detection of multiple objects and the assessment of the physiological state of marine microalgae.

“In this study, we trained one-stage object detection algorithms, including YOLOv5, YOLOv8, and TOOD, and two-stage object detection algorithms, including Faster-RCNN, Cascade-RCNN, and Dynamic-RCNN,” the researchers report.

Microalgae Image Database

The study authors established a database for the detection of marine microalgae in microscopy images, including collection data, microscopy images, etc.

See also  Electrochemical Applications in Recirculating Aquaculture Systems

The database contains 967 images of microalgae from six genera: Pinnularia, Chlorella, Platymonas, D. salina, Isochrysis, and Symbiodinium. In addition, they also collected images of Symbiodinium in different physiological states.

Conclusion

“The results showed that one-stage and two-stage object detection models can achieve high average accuracy, demonstrating the ability of computer vision in detecting multiple microalgae objects and providing basic data and models for detection in real-time from microalgae cells”, they conclude.

The study was funded by the Hainan Provincial Key Research and Development Program, the Hainan Provincial Higher Education Scientific Research Project Key Project, the Hainan Provincial Social Science Research Project, and the Hainan Provincial Natural Science Foundation.

Reference (open access):
Zhou, S., Jiang, J., Hong, X., Fu, P., & Yan, H. (2023). Vision meets algae: A novel way for microalgae recognization and health monitor. Frontiers in Marine Science.

Leave a Comment