Inkjet printer ink drop feature detection based on machine vision

Continuous inkjet printer is an indispensable equipment in product life cycle traceability management. In order to ensure the quality of printing, it is necessary to analyze the changes of ink droplets in the process of spraying. Firstly, an inkdrop detection system based on a telecentric industrial camera was established to collect inkjet printer images under 150, 170 and 200 pressure conditions. Secondly, the ink drops are preprocessed to remove noise and refine the edge of the ink drop shape, and then the ink drop filling images under various edge detection operators are compared to obtain the best ink drop filling image and contour texture image. Finally, the characteristic parameters such as circularity, complexity, width to length ratio and eccentricity of the ink drop were extracted. The experimental results demonstrate that the morphological changes of ink droplets during the injection process can be comprehended through analysis of the variation curves of characteristic parameters under different pressure conditions, enabling detection of adhesion, trailing, and satellite ink droplets. The research results have important guiding significance for improving the printing quality of continuous inkjet printer.