Automating fish sorting

Norwegian scientists have developed a program that can analyse and sort fish according to quality.

Ekrem Misimi, from SINTEF Fisheries and Aquaculture Research, has combined machine vision with pattern recognition methods and has fed geometrical descriptions of the size, colour and shape of salmon into a PC, which then grades the fish according to its quality.

Today, employees manually grade fish by assessing their shape, colour and surface injuries. They also watch for flecks of blood on the flesh (caused by blood in the stomach cavity) and for rigor mortis.

The new method simply compares the colour of a fillet to stored values from a colour table. This rapid method is similar to current colour analysis and does not require physical contact with the fish.

"Machine vision and image analysis will enable us to sort fish into 'production', 'ordinary' and 'superior' classes, while revealing blood in the stomach cavity, with an accuracy of 90%," said Misimi.