A Method of Binocular Laser 3-D Scanning Imaging for Reflective Workpieces

A deep learning line laser 3-D measurement method based on feature fusion and attention mechanism is proposed to address the impact of reflective workpieces on the extraction of laser stripe centers. First, a UNet segmentation model based on feature fusion and attention mechanism is established. The deep learning model can effectively solve the interference caused by reflection and can segment the overall distribution and bending characteristics of laser stripes. Second, the Steger algorithm is used to roughly extract the center of the laser stripe, and the contour polygon segmentation method is used to adaptively obtain the segmentation points of the laser stripe. Finally, polynomial fitting is performed based on segmented points to obtain a smoother laser stripe centerline. The experimental results show that the proposed method can effectively overcome the interference caused by reflective workpieces and generate a smoother 3-D model, and the measurement repeatability error is less than 0.37 mm, and the relative error is less than 0.03 mm.

Automated summary

A deep learning line laser 3-D measurement method based on feature fusion and attention mechanism is proposed to address the impact of reflected workpieces. A UNet segmentation model is established to solve the interference caused by reflection and segment the overall distribution and bending characteristics of laser stripes. The Steger algorithm is used to extract the center of the laser stripe, and the contour polygon segmentation method is used to obtain the segmentation points of the laser stripe. Polynomial fitting is then performed to obtain a smoother laser stripe centerline. The proposed method effectively overcomes interference and generates a smoother 3-D model.