DF Scanner
Desktop application for forest information analysis
DF Scanner
Desktop application for forest information analysis
DF Scanner is software designed to analyze forest information, including tree species, size, and timber volume, from data acquired by drones in the forest.
It is primarily designed for tree species identification using deep learning from images of tree canopies captured by drones. It also includes various analyses at the individual tree level, such as CHM (Canopy Height Model) calculation, tree crown detection, crown separation, and estimation of DBH (Diameter at Breast Height), timber volume, and carbon storage.
Tree top detection tree crown segmentation
It is possible to detect the tree canopy tops and separate tree crowns of coniferous trees using height and image information. The algorithm is based on the use of a Local Maximum Filter and Watershed Segmentation. By adjusting multiple parameters, such as filter size, it is possible to achieve a detection accuracy of over 90% in coniferous forests.
Tree species identification
AI-based tree species identification is possible using drone image data, leveraging deep learning. It is not limited to the automatic identification of certain coniferous and broadleaf tree species found in Japan but also supports user-driven training and identification with custom classes.
For user-driven training and identification, the software offers features that are low-load and can be run on a CPU, such as site tuning classification, one-time deep learning for single-session training and identification, and my deep learning, which allows you to create original models from multiple data sources.
Users can also configure hyperparameters like batch size and epochs to fine-tune the process.
Size estimation
The software allows for the estimation of tree height, diameter at breast height (DBH), and timber volume.
The estimation of DBH is calculated based on tree species, tree crown (polygon) size, and tree height information. Timber volume estimation takes into account tree species, tree height, DBH, and regional factors.
Additionally, it is possible to customize size estimations based on locally measured data using proprietary formulas.
Understanding the forest condition
You can view information such as the selected area’s tree species, count, average size, total timber volume, and more. The software also supports the loading of shapefiles for the selected area, making it easy to assess forest information for specific regions.
Product Price
DF Scanner
Subscription License
Subscription License
Domestic Price (Japan), including tax
DF Scanner
Perpetual License
Perpetual License
Domestic Price (Japan), including tax
Only Lite version
JPY 880,000
（Annual maintenance fee after the first year: JPY 88,000 per year）
For Overseas Pricing
Please contact us.
Pro vs Lite Feature Comparison Table
| Features | DF Scanner Pro | DF Scanner Lite |
|---|---|---|
| Data Visualization | ○ | ○ |
| Fully Automated Analysis | ○ | ○ |
| Coniferous Crown Segmentation | ○ | ○ |
| Broadleaf Crown Segmentation | ○ | × |
| AI Species Classification (Cedar & Cypress) | ○ | ○ |
| AI Species Classification (Hokkaido Conifers) | ○ | ○ |
| AI Species Classification (Japanese Tree Species) | ○ | × |
| AI Dead Tree Detection | ○ | × |
| AI Quick Learning / Classification (Site-Tuning) | ○ | ○ |
| AI Model Training / Classification | ○ | × |
| Tree Height & DBH Estimation | ○ | ○ |
| Log Volume Estimation | ○ | ○ |
| CO₂ Sequestration Estimation | ○ | ○ |
| Statistics Output | ○ | ○ |
| Project Save *1 | ○ | ○ |
| Online Backup *1 | ○ | ○ |
*1 Not available in the free trial version.
DF Scanner System Requirements
| Required | Recommended | |
| OS | Windows10,11 | – |
| RAM | 4GB | 16GB or more |
| Free Disk Space | 7GB | 10GB or more |
| GPU | – | NVIDIA GeForce RTX series (Some older models may not be compatible.） |
*1 For tree species identification with DF Scanner, using a graphics board equipped with an NVIDIA GPU that supports CUDA 12.8 enables high-speed processing. For a list of compatible GPUs, please click here.
CONTACT
