Method for detecting surface defects of underwater buildings: Binocular vision based on sinusoidal grating fringe assistance

Abstract

In underwater building-surface defect detection, high pressure, cold and other underwater factors will cause harm to frogmen, sonar and other detection techniques contain limited information and accuracy is difficult to meet the demand. In view of the above disadvantages, a sine stripe-assisted visual detection method of underwater building surface-defects is proposed. By generating a fixed sine stripe by program and projected it on the surface of the underwater building, and then control a binocular CCD to acquire six frames of sinusoidal stripes with different phases in one sampling period. Experiment results show that the error of our method does not exceed 0.5 mm and the reconstruction efficiency is not less than 5 m2/s in the 3D dimensional measurement of the defects on the surface of the submerged building body.

Keywords

• Underwater buildings
• Surface defects
• Sinusoidal grating stripes
• Binocular vision
• 3D point cloud reconstruction