One promising approach to meet the increasing protein demand of the world population is rearing insects on organic substrates. An insect processing plant's production of alternative organic protein leads to a circular economy that recovers nutrients and minimises waste volume. However, there is still a need for further automation of the production plants to reduce operating costs. This master thesis details the setup and the performed trials of a neonate counting system implemented at an insect processing plant. The application is based on an optical counting approach using a line scan camera. This thesis evaluates the setup between the camera unit and the feeding unit to achieve good image quality for evaluating the recorded images. For this purpose, trials cover ideal particles (precision balls) within a laboratory setting and real particles (neonates) at an insect processing plant. The results of the performed trials detail that the presented system is suitable for neonate counting. The evaluation of the trials results in a relative counting error of 0,20 % using ideal particles (precision balls) and 7,19 % using real particles (neonates), whereby the main counting error can be attributed to insufficient detection of neonate agglomerates. Counting neonates with a line scan camera yields good results despite the neonates' particle characteristics.