Design and simulation of plenoptic imaging system on unmanned aerial vehicle for 3D aerial photography
3D imaging by aerial photography is an important tool for environmental monitoring, infrastructure inspections, etc. The plenoptic imaging system, composed of a micro-lens array, acquires multiple images with various view angles simultaneously and thus is more accurate than traditional stereo vision systems, for precise baseline length in stereo matching and subpixel algorithm of multiple images with pixel shifts. Moreover, in dark environments, lighting from an unmanned aerial vehicle may cause specular reflection from the surface of targets and disturbs height detection. The plenoptic imaging system for 3D aerial photography is designed with working distance of about 3m, field of view of 1.5m x 1.1m, lateral resolution of 0.5mm, and height resolution of 0.3m. The design is modeled in Blender software and simulated with various lighting conditions. With a small light source close to the plenoptic imaging system, detected heights at the region with obvious specular reflection are distorted. On the other hand, with an area light source close to the surface of targets, detected heights are more accurate. The lighting environment can be achieved by focusing an area light source on the surface of targets by an external optical system. Generally an internal coaxial lighting infrastructure is applied to avoid mechanical interference, and the detailed optical system with an internal coaxial light source will be designed and simulated in the future.
Authors
- Ming-Fu Chen (ORCID: https://orcid.org/0009-0005-7016-2210)
- X. Liu (ORCID: https://orcid.org/0009-0000-4407-0316)
- Chih-Hao Lin
Institutions
- Taiwan Forestry Research Institute (TW)
Publication Details
- Published
- 2025-01-10
- DOI
- https://doi.org/10.1117/12.3042514
- Primary Topic
- Advanced Measurement and Detection Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00