A new method of piezoelectric inkjet ink drop recognition based on machine vision

In the field of 3D printed circuit board, the requirements for piezoelectric inkjet ink droplets are getting higher and higher. The traditional ink droplet observation method is not perfect and the accuracy is unstable. Therefore, this paper studies the identification and observation of piezoelectric inkjet ink droplets based on the non-contact principle of machine vision. Firstly, a telecentric machine vision measurement system is built to collect effective ink drop images. Then combined with sub-pixel edge extraction operator and least square method, an improved edge least square fitting method is designed and compared with regional least circumcircle method and edge least circumcircle method. The experimental results show that the ink droplet radius obtained by the method proposed in this paper has small error and high precision. The results obtained by calculating the radius of the edge are more accurate, and can be used to monitor the existence of satellite ink droplets. The research results have important guiding significance for improving the circuit inkjet effect.