Investigation on Wetting of Microdroplets Enhancement on Glass Dent for Industrial Inspection

Dents have a significant impact on the quality of glass covers, but the detection of surface dents has been a difficult task due to the unique optical properties of glass. In order to reduce the difficulty of glass dent detection, this paper proposes an image enhancement method based on droplet wetting. It is shown that by spraying the surface of the glass cover on the hydrophobic side through a nebulizer, the location of the defects can be directly identified under a simple optical scheme. Thanks to this enhancement, the subsequent image detection method becomes more concise and clear. In this paper, the motion of the droplets falling near the defects during the spraying process is investigated by hydrodynamic simulation, and the imaging enhancement phenomenon is explored in combination with optical simulation and actual imaging experiments. The experimental results are consistent with the simulation results, indicating that the model and its mechanism are feasible. The proposed method improves the detection accuracy of glass defects by an average of 44% in spatial- and frequency-domain algorithms and by 38% in neural-network-based approaches.