Bark-Code
Biometric Tree Identification Through LiDAR Bark Morphology Analysis
Duration
2024 – 2025
Researcher
Hanju Seo
Affiliation
- Imperial College London · Dyson School of Design Engineering
- Royal College of Art · School of Design
Role
- Principal Investigator
- Hardware Designer (LiDAR scanner prototype)
- UX/UI Designer (scanning interface & user flow)
- Field Researcher (Finland, UK, South Korea)
Conference
- Dubai AI Festival, Dubai, 2025
- Grantham Institute Climate Research Showcase, London, 2025
- Imperial Futures Poster Exhibition, London, 2025
- NEST Conference, University of Sussex, Brighton, 2025
Fund
- D&AD Future Impact Fund 2025 - £25,000
Award
- Dezeen x Bentley Start Something Powerful Award - £15,000
- Seoul Design Award 2025 - Best of the Best (KRW 5,000,000)
- Core77 Design Awards 2025 - Emerging Technologies, Student Winner
- D&AD Awards 2025 - Future Impact Shortlist
- WIRED Creative Hack Award 2024 - Finalist
Bark-Code began as an attempt to establish a shared language between nature and machine: a codification of the living world that does not require altering it. Every tree carries on its surface a structurally unique texture, formed by the arrangement of lenticels, the spacing of ridges, and the irregular grain of bark developed over years of growth. Bark-Code treats this texture as a biometric signature, readable by a handheld LiDAR scanner and traceable across time without any physical tag or invasive marker.
The research addressed a verification gap that has long limited smallholder participation in carbon credit markets. Family forest owners lack access to affordable, reliable tools for documenting their trees. Without verifiable, persistent identity for each stem, their forests cannot participate in the data infrastructures that underpin carbon accounting.
Field research was conducted across Finland, the United Kingdom, and South Korea, with direct interviews carried out with small and family forest owners in each country. Their recurring constraints shaped every design decision, from the hardware form factor to the scanning interface. The hardware prototype was designed to be handheld, low-power, and operable without specialist training. A proof-of-concept was developed using LiDAR-based point cloud capture combined with morphological feature extraction, demonstrating that individual trees could be re-identified across multiple scanning sessions with high accuracy. The project was presented at the Dubai AI Festival, where the platform was demonstrated to an international audience spanning agritech, climate finance, and AI research communities.
Field research, hardware prototyping, and exhibition documentation across South Korea, Finland, and the United Kingdom.