dislocation-density

A reproducible batch processing program for estimating dislocation density from Transmission Electron Microscopy (TEM) images, implemented as an M1–M8 modular pipeline: IO & Calibration, Preprocessing, Line Enhancement, Segmentation, Skeletonization, Metrics, QA Visualization, CLI & Configuration.

Installation

Recommended to use a virtual environment, then install from the root directory:

python -m pip install -U pip
python -m pip install -e .

Quick Start

1. Copy config:

cp config.example.yaml config.yaml

2. Run batch processing:

dislo_density run --input ./ --out results/ --config config.yaml

Web Interface (Flask)

Launch local web interface:

python -m pip install -e .
dislo_density_web --config config.yaml --host 127.0.0.1 --port 5000

Browser: http://127.0.0.1:5000/

Output

Generates in results/:

• summary.csv: Summary results per image/ROI
• summary_stats.csv: Mean/Std statistics
• results/<image_stem>/result.json: Detailed JSON results
• results/<image_stem>/roi_XX/overlay.png: Skeleton overlay
• results/<image_stem>/roi_XX/qc.png: Quality control panel

Calibration (nm/px)

Priority:

1. m1.manual_nm_per_px: Manual input (Recommended)
2. TIFF metadata
3. Scale bar detection + Manual value (OCR removed)