The Narlu Case Study is the start of an ongoing project on monitoring the rehabilitation of a degraded dairy farm back into rainforest in the Northern NSW hinterland. The owners, the Vercoe family are passionate about rehabilitating the property and had already collected very useful information. In addition, Kaluta gathered data on the ground and from the air, combining it to generate 2D maps, 3D models and a considerable amount of data on the 34Ha property to showcase it as it was in June 2021 near the beginning of the project.
Tools used were drones with both RGB and thermal imaging, photogrammetry software, GPS and visual observations. Back at the desktop the data from these were refined. The outcomes were:
- Digital terrain models
- LiDAR compatible modelling
- 3D OBJ files
From the file type outcomes listed above, we were able to extract:
- Small weeds using a trained algorithm and specialised workflow
- Contours from the digital terrain models
- Weed counts/cover from the orthomosaics
- Native plant cover from the orthomosaics
- Drainage trends from the contours and orthomosiacs
- Plant health from the trained algorithm and thermal imagery
- Physical measurements of the property from the orthomosaics
- Planting site counts from the orthomosaics
- Density maps from the orthomosaics
Now that the important information has been converted to numbers and Key Performance Indicators have been outlined, the success can be easily tracked over time and scored, allowing us to view the progress of this project.
The workflows and techniques explored and refined over the course of this case study are easily able to be used on other projects.
Included is a bonus section on thermal animal surveys and thermal image analysis.