tree-disk-pith 0.1.2

A package for tree disk pith detection in images

Tree Disk Pith Detection

A Python package for analyzing tree rings in cross-sectional images. Originally forked from hmarichal93/apd.

Installation

pip install tree-disk-pith

Usage

Python API

import treediskpith

# Configure the analyzer
treediskpith.configure(
    input_image="input/tree-disk4.png",
    save_results=True,
)

# Run the detection
(
    img_in,          # Original input image
    img_pre,         # Preprocessed image
    pith,  # Center of the tree disk
) = treediskpith.run()

Command Line Interface (CLI)

Basic usage:

tree-disk-pith --input_image ./input/tree-disk3.png --new_shape 640 --debug

Save intermediate results:

tree-disk-pith --input_image ./input/tree-disk3.png --new_shape 640 --debug --method apd_pcl --save_results

Advanced usage with custom parameters:

tree-disk-pith \
    --input_image input/tree-disk3.png \
    --cx 1204 \
    --cy 1264 \
    --output_dir custom_output/ \
    --sigma 4.0 \
    --th_low 10 \
    --th_high 25 \
    --save_results \
    --debug

CLI Arguments

Argument	Type	Required	Default	Description
--input_image	str	Yes	-	Input image file path
--output_dir	str	Yes	-	Output directory path
--method	str	No	apd	Detection method to use. Choices are apd, apd_pcl, or apd_dl
--model_path	str	No	-	Path to the weights file (required if using apd_dl method)
--percent_lo	float	No	0.7	percent_lo parameter for the algorithm
--st_w	int	No	3	st_w parameter for the algorithm
--lo_w	int	No	3	lo_w parameter for the algorithm
--st_sigma	float	No	1.2	st_sigma parameter for the algorithm
--new_shape	int	No	0	New shape for resizing the input image. If 0, no resizing is done
--debug	flag	No	False	Enable debug mode to save intermediate images and outputs
--save_results	flag	No	False	Save intermediate images, labelme and config file

Development

Setting up Development Environment

1. Clone the repository:

git clone https://github.com/tuke307/tree-disk-pith.git
cd tree-disk-pith

2. Create and activate virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install development dependencies:

pip install -r requirements.txt

4. Install the package in editable mode:

pip install -e .

5. fetch dataset

python fetch_dataset.py

6. Download pretrained model

python fetch_pretrained_model.py