TRAS

Tree Ring Analyzer Suite

Professional dendrochronology software integrating state-of-the-art computer vision and deep learning methods for automatic tree ring detection and measurement

Download Latest View on GitHub Documentation Tutorial

Overview

TRAS is a comprehensive GUI application built with PyQt5 that combines three state-of-the-art methods for dendrochronology research: automatic pith detection (APD), classical edge detection (CS-TRD), and deep learning-based detection (DeepCS-TRD).

Reduces analysis time from hours to minutes

State-of-the-art deep learning models

Professional PDF reports & exports

Used by researchers worldwide

Key Features

🔬 Automatic Detection

APD: Automatic Pith Detection (~1s, CPU)
CS-TRD: Classical edge detection (~73s, CPU) [Linux/macOS]
DeepCS-TRD: Deep learning U-Net (~101s, GPU)

🖼️ Image Preprocessing

Crop: Focus on region of interest
Resize: Optimize for detection algorithms
Background Removal: U2Net-based intelligent removal

📏 Scale Calibration

Convert pixel measurements to physical units
Support for mm, cm, and μm
Visual scale bar display

📊 Analysis & Export

CSV: Ring properties with physical measurements
.POS: CooRecorder format for dendrochronology
JSON: Complete annotation persistence

📄 Professional PDF Reports

Generate comprehensive, publication-ready reports with a single click. Each report includes sample metadata, ring overlays, and multi-panel analysis plots.

Cover Page

Metadata & statistics

Ring Overlay

Detected boundaries & labels

Analysis Plots

Area, growth rate, radial width

Installation

TRAS requires Anaconda/Miniconda for dependency management. Windows installation tutorial

Step 1: Download TRAS

Download manually from Latest Release and extract

Important Note for Windows Users

After downloading and extracting TRAS, you need to:

Open Anaconda Prompt (not regular Command Prompt - search for "Anaconda Prompt" in the Start menu)
Navigate to the extracted directory: cd path\to\tras-2.0.4
Then follow the installation commands below

Step 2: Install with Conda

Create Conda Environment

conda env create -f environment.yml

Activate Environment

conda activate tras

Install TRAS

pip install -e .

Download models

python tools/download_release_assets.py

Step 3: Additional Components (Optional)

Compile CS-TRD (Linux/macOS only)

# From the tras-2.0.4 directory
cd tras/tree_ring_methods/cstrd/devernay && make clean && make &&
cd ../../..  # Return to tras-2.0.4 directory

Quick Start

Launch TRAS and follow the guided workflow to analyze your tree ring samples:

Activate Environment in Anaconda Prompt

conda activate tras

Launch Application

tras

Load Image

Open your wood cross-section image (JPG, PNG, TIFF supported)

Set Metadata

Enter sample code, harvested year, and observations

Calibrate Scale

Draw a reference line and specify its physical length

Preprocess (Optional)

Crop, resize, or remove background to improve detection

Detect Rings

Choose APD for pith, then CS-TRD or DeepCS-TRD for rings

Manual Editing (Optional)

Add/remove points, create new rings, or delete false detections

Measure Radial Width

Set direction from pith and export to .POS format

Analyze & Export

View ring properties, generate PDF reports, export CSV data

View Full Tutorial

Citations

If you use TRAS in your research, please cite the relevant papers:

📘 Automatic Pith Detection (APD)

@inproceedings{apd, isbn = {978-3-031-78447-7}, year = {2025}, pages = {1--15}, title = {Automatic Wood Pith Detector: Local Orientation Estimation and Robust Accumulation}, author = {Marichal, Henry and Passarella, Diego and Randall, Gregory}, abstract = {A fully automated technique for wood pith detection (APD), relying on the concentric shape of the structure of wood ring slices, is introduced...}, booktitle = {International Conference on Pattern Recognition (ICPR)} }

View Paper View Code

📘 Classical Tree Ring Detection (CS-TRD)

@misc{marichal2024cstrd, title={CS-TRD: a Cross Sections Tree Ring Detection method}, author={Henry Marichal and Diego Passarella and Gregory Randall}, year={2024}, eprint={2305.10809}, archivePrefix={arXiv}, primaryClass={cs.CV}, url={https://arxiv.org/abs/2305.10809}, }

arXiv Paper IPOL Article

📘 Deep Learning Detection (DeepCS-TRD)

@misc{marichal2025deepcstrd, title={DeepCS-TRD, a Deep Learning-based Cross-Section Tree Ring Detector}, author={Henry Marichal and Verónica Casaravilla and Candice Power and Karolain Mello and Joaquín Mazarino and Christine Lucas and Ludmila Profumo and Diego Passarella and Gregory Randall}, year={2025}, eprint={2504.16242}, archivePrefix={arXiv}, primaryClass={cs.CV}, url={https://arxiv.org/abs/2504.16242}, }

arXiv Paper Project Page

Contact

Have a question about TRAS, need support, or want to collaborate on dendrochronology projects? Reach out through my email.

General Questions

hmarichal93@gmail.com

Stay connected for release notes, datasets, and paper announcements: