Commercial tuna fishing generates more than US$40 billion in economic activity each year. A single mature Pacific bluefin tuna can command a price in excess of US$1.5million. Recently, as part of a drive towards more sustainable management of fish stocks, fishing vessels have been required to accurately count and report the number of young bluefin tuna caught in the wild – where they go on to be used for breeding at fish farms.
Counting fish is also necessary in the aquaculture industry where it is essential to know the number of fish inside a holding net in order to control production volume and the amount of feed used.
New automated fish counting system with 98% accuracy
To address both needs, Japanese company Yanmar Marine Systems Co., Ltd. (YMS) has begun accepting local orders for its new automated fish counting system.
Yanmar’s Research & Development Center developed image recognition and processing technology and an integrated system including a specially designed underwater camera and image processing computer, to realize real-time automated counting. Under optimal conditions, the system has succeeded in automatically counting tuna with an accuracy of more than 98%.
In the future, Yanmar aims to contribute to greater efficiency and development of the fishing and aquaculture industry by reducing feed costs, cutting work time, and facilitating the accurate reporting of catches. Existing manual methods of counting fish are laborious and time-intensive requiring the manual counting of the total number of fish caught using underwater images.
The system enables users to analyse and check results during or immediately after video capture.
- Angle of view of underwater camera can be adjusted by remote control even after the camera is installed underwater, reducing the time required to install and adjust the camera.
- Brightness can be adjusted to reduce image disturbances and reduce impact on automated analysis.
- Screen display and correction functionality enables easy manual counting and correction after automated analysis even for images affected by environmental disturbances. * If the image is affected by environmental disturbance, counting correction is required after the automated analysis process.
