Enumeration, measurement, and identification of net zooplankton samples using the ZOOSCAN digital imaging system

Grosjean, P., Picheral, M., Warembourg, C., and Gorsky, G. 2004. Enumeration,measurement, and identification of net zooplankton samples using the ZOOSCAN digitalimaging system. e ICES Journal of Marine Science, 61: 518e525.Identifying and counting zooplankton are labour-intensive and time-consuming processesthat are still performed manually. However, a new system, known as ZOOSCAN, has beendesigned for counting zooplankton net samples. We describe image-processing and theresults of (semi)-automatic identification of taxa with various machine-learning methods.Each scan contains between 1500 and 2000 individuals !0.5 mm. We used two trainingsets of about 1000 objects each divided into 8 (simplified) and 29 groups (detailed),respectively. The new discriminant vector forest algorithm, which is one of the mostefficient methods, discriminates between the organisms in the detailed training set with anaccuracy of 75% at a speed of 2000 items per second. A supplementary algorithm tagsobjects that the method classified with low accuracy (suspect items), such that they could bechecked by taxonomists. This complementary and interactive semi-automatic processcombines both computer speed and the ability to detect variations in proportions and greylevels with the human skills to discriminate animals on the basis of small details, such aspresence/absence or number of appendages. After this checking process, total accuracyincreases to between 80% and 85%. We discuss the potential of the system as a standard foridentification, enumeration, and size frequency distribution of net-collected zooplankton.