Edge Linking Based Method to Detect and Separate Individual C. Elegans Worms in Culture

Nowadays an increasing research interest in the field of biotechnology has been drawn to achieve reliable information from model organisms. C. Elegans nematode worm is one of the major animals. Machine vision analysis of this animal needs to solve many important problems, e.g. detection of each individual in population images, movement patterns of isolated and overlapped worms and so on. In this paper, we describe our recently proposed method with an analysis to show the impact of two major parameters on the method efficiency. Based on our analysis on 255 isolated and overlapped worms, we find that our method prepares the best correct detection (TAR=83%) for thetas
_Th
and N
_Th
between 15 and 20 pixels.