TRAS

Tree Ring Analyzer Suite

Version 2.0.0

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

Release Downloads License Python Platform
Download v2.0.0 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).

TRAS Main Interface
Reduces analysis time from hours to minutes
State-of-the-art deep learning models
Professional PDF reports & exports
Used by researchers worldwide

Key Features

Detection Methods

🔬 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

🖼️ Image Preprocessing

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

📏 Scale Calibration

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

📊 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.

PDF Report Page 1

Cover Page

Metadata & statistics

PDF Report Page 2

Ring Overlay

Detected boundaries & labels

PDF Report Page 3

Analysis Plots

Area, growth rate, radial width

Installation

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

Step 1: Download TRAS v2.0.0

Linux / macOS

wget https://github.com/hmarichal93/tras/archive/refs/tags/v2.0.0.tar.gz tar -xzf v2.0.0.tar.gz cd tras-2.0.0

Windows (PowerShell)

Invoke-WebRequest -Uri "https://github.com/hmarichal93/tras/archive/refs/tags/v2.0.0.zip" -OutFile "v2.0.0.zip" Expand-Archive -Path "v2.0.0.zip" -DestinationPath . cd tras-2.0.0

Alternative: Download manually from Releases and extract

Important Note for Windows Users

After downloading and extracting TRAS, you need to:

  1. Open Anaconda Prompt (not regular Command Prompt - search for "Anaconda Prompt" in the Start menu)
  2. Navigate to the extracted directory: cd path\to\tras-2.0.0
  3. 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 .

Step 3: Install Urudendro

Urudendro is a utility library required for background removal (U2Net) and DeepCS-TRD functionality.

Install from GitHub

pip install git+https://github.com/hmarichal93/uruDendro.git

Note: Requires git-lfs to be installed on your system.

Step 4: Additional Components (Optional)

Compile CS-TRD (Linux/macOS only)

# From the tras-2.0.0 directory cd tras/tree_ring_methods/cstrd/devernay make cd ../../../.. # Return to tras-2.0.0 directory

Note: CS-TRD is not available on Windows. Windows users should use DeepCS-TRD.

Download DeepCS-TRD Models (Optional)

Linux/macOS:

# From the tras-2.0.0 directory cd tras/tree_ring_methods/deepcstrd ./download_models.sh cd ../../.. # Return to tras-2.0.0 directory

Windows:

# From the tras-2.0.0 directory cd tras\tree_ring_methods\deepcstrd python -c "import urllib.request; import zipfile; import os; url='https://github.com/hmarichal93/tras/releases/download/models/deepcstrd_models.zip'; urllib.request.urlretrieve(url, 'models.zip'); zipfile.ZipFile('models.zip').extractall('models'); os.remove('models.zip')" cd ..\..\.. # Return to tras-2.0.0 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