Triple line-scan camera measurement for efficient and accurate narrow-space 3-D scanning with motion estimation

Area of Science:

• Computer Vision
• Metrology
• Transportation Engineering

Background:

• Vision-based 3-D shape measurement is vital for inspecting vehicles and trains.
• Conventional methods face challenges in narrow-space inspection.
• Increasing demand for comprehensive narrow-space inspection necessitates advanced solutions.

Purpose of the Study:

• To propose an efficient and accurate narrow-space 3-D scanning method.
• To overcome limitations of conventional vision methods in confined environments.
• To enable detailed 3-D shape acquisition for transportation infrastructure.

Main Methods:

• Utilizing a novel triple line-scan camera structure.
• Employing coplanar dual line-scan cameras with a narrow reflector for optical path access.
• Integrating an additional line-scan camera for time-space constraints in point cloud stitching.
• Implementing an efficient algorithm template for matching and motion estimation.
• Performing point cloud texture mapping and light source optimization.

Main Results:

• Achieved efficient in-motion measurement with a large field of view in narrow spaces.
• Ensured accurate point cloud stitching through the triple line-scan camera structure and algorithms.
• Demonstrated low-distortion, in-motion 3-D reconstruction despite motion variations.
• Successfully obtained 3-D shape information in challenging narrow environments.

Conclusions:

• The proposed triple line-scan camera method is effective for narrow-space 3-D shape measurement.
• The system overcomes conventional limitations, enabling efficient and accurate inspection.
• The technique is suitable for in-motion reconstruction in dynamic, confined settings relevant to transportation.