In order to accurately detect the abnormal looseness of strapping in the process of steel coil hoisting, an accurate detection method of strapping abnormality based on CCD structured light active imaging is proposed. Firstly, a maximum entropy laser stripe automatic segmentation model integrating multi-scale saliency features is constructed. With the help of saliency detection model, the purpose is to reduce the interference of the environment to the laser stripe and highlight the distinguishability between the stripe and the background. Then, the maximum entropy is used to segment the fused saliency features and accurately extract the stripe contour. Finally, the stripe normal field is obtained by calculating the stripe gradient vector, the stripe center line is extracted based on the stripe distribution normal direction, and the abnormal strapping is recognized online according to the stripe center. Experiments show that the proposed method is effective in terms of detection accuracy and time efficiency, and has certain engineering application value.

Ambient noise and illumination inhomogeneity will seriously affect the high-precision measurement of structured light 3D morphology. To overcome the influences of these factors, a new, to the best of our knowledge, sub-pixel extraction method for the center of laser stripes is proposed. First, an automatic segmentation model of structured light stripe based on the UNet deep learning network and level set is constructed. Coarse segmentation of laser stripes using the UNet network can effectively segment more complex scenes and automatically obtain a prior shape information. Then, the prior information is used as a shape constraint for fine segmentation of the level set, and the energy function of the level set is improved. Finally, the stripe normal field is obtained by calculating the stripe gradient vector, and the center of the stripe is extracted by fusing the gray center of gravity method according to the normal direction of the stripe distribution. The experimental results show that the average width error of different rows of point cloud data of workpieces with different widths is less than 0.3 mm, and the average repeatability extraction error is less than 0.2 mm.

No abstract