Abstract: As a kind of in vivo model organism, C. elegans（hereinafter referred to as nematode） is widely used in drug screening and neurobiology research. Physiological index（body length, etc.） is an important part of nematode research. The use of image recognition to study the morphological characteristics of nematodes has higher precision than traditional methods. In recent years, microfluidic chips have been increasingly used in the study of cells and microscale organisms due to their characteristic scale matching with nematode size scale and easy automation. This paper introduces a nematode image recognition system based on microfluidic chip. The system uses the designed microfluidic chip to realize the rapid flow of single-worms through the detection channel, and builds a machine-based machine vision module based on machine learning. The morphological characteristics of nematodes were quickly classified, and the classification results could be used as the basis for subsequent sorting operations.