Case Study
Ash Dieback Assessment & AI-Powered Tree Survey
Client : National Highways (pilot project)

The Challenge
Managing trees alongside the strategic road network requires a careful balance between safety, network availability and environmental responsibility.
Traditional ground-based tree surveys can be time-consuming and difficult where access is restricted by live traffic, embankments or dense vegetation. Ash Dieback adds another challenge, with affected trees requiring effective monitoring to help identify potential risks and prioritise further inspection.
The A34 pilot explored how drone technology, LiDAR and artificial intelligence could provide a faster, more consistent way to understand tree condition across a highway corridor.
From Survey Data to Asset Intelligence
Rather than simply producing aerial imagery, VM-Ai transformed the captured data into structured, usable vegetation intelligence.
The survey provided:
- Individual tree detection and mapping
- Identification of likely Ash trees
- Tree height and crown measurements
- Canopy loss assessment
- Tree-level location and condition records
- GIS-ready data for future monitoring and comparison
Canopy loss was grouped into consistent bands, helping provide a screening tool to identify trees that may require closer review by qualified arboricultural specialists.
The VM-Ai Solution
Direct Enviro Services used drone-mounted LiDAR and high-resolution aerial imagery, combined with VM-Ai processing, to create a detailed digital representation of the survey area.
AI analysis was then used to identify and separate individual trees, recognise likely Ash specimens and assess key characteristics including tree height, crown extent, maturity and estimated canopy loss.
Each tree could be treated as an individual digital asset, with its location and survey information prepared for integration into GIS.
Supporting Safer, More Targeted Management
The pilot demonstrated how remote surveying can help teams understand large and difficult-to-access areas before sending inspectors into the field.
By identifying potential areas of concern remotely, specialist inspections can be targeted where they are needed most. This has the potential to reduce unnecessary roadside exposure, focus traffic management and specialist access requirements, and provide a consistent digital record for future surveys.
Better Data. Smarter Decisions
The A34 trial demonstrated how drones, LiDAR, high-resolution imagery and AI can work together to support more data-led highway vegetation management.
VM-Ai does not replace professional arboricultural assessment. It gives specialists a much broader set of eyes, helping them understand where trees are, identify potential concerns and prioritise where their expertise is needed most.





