PCB ReTrace

Digitize, document, and reverse engineer printed circuit boards entirely in your browser.

Launch PCB ReTrace

🔍 Overview

PCB ReTrace is a local-first web application designed for electronics engineers, repair technicians, and hardware hackers. It helps you move from a physical PCB to a digital representation (BOM & Netlist) using only photographs.

It runs 100% client-side using HTML5, Vanilla JS, and WebAssembly. No data is ever uploaded to a server.

✨ Key Features

BOM Management: Map components on board photos to a digital Bill of Materials.
Visual Inspection: View multiple board layers (Top/Bottom) simultaneously.
Computer Vision Stitching: Automatically align close-up macro shots with wide-angle reference photos using OpenCV.
X-Ray View: Manually stitch Top and Bottom images to trace vias and through-holes across layers.
Netlist Tracing: click-to-trace connectivity and export to KiCad compatible netlists (.net).
Integrated Tools:
- Resistor Color Code Decoder (3-6 bands)
- Inductor Color Code Decoder
- Air-core Coil Calculator
Privacy First: All data is stored in your browser's IndexedDB.

🚀 Getting Started

Online

Simply visit the GitHub Pages deployment.

Offline / Local Development

Since the project uses no build tools, you can run it directly from the source:

Clone this repository:

git clone https://github.com/dev-lab/pcb-retrace.git

Navigate to the docs folder (where the source code lives):
```
cd pcb-retrace/docs
```
Open index.html in your browser.
- Note: Some features (like Camera access) require a secure context (HTTPS) or localhost. For best results, use a simple local server like Python's http.server.

📂 Project Structure

This project follows a "Build-Free" architecture. All source code is located in the /docs folder to support GitHub Pages directly.

docs/studio.html: The main application entry point.
docs/studio.js: Core controller and logic.
docs/cv-core.js: Computer Vision & Homography logic (OpenCV wrapper).
docs/inspector.js: Visual trace tracking logic.
docs/nets.js: Netlist management and KiCad export.

⚖️ License & Commercial Use

This project is Dual Licensed:

Open Source (AGPLv3): Ideal for hobbyists, educational use, and open-source projects. You are free to use and modify the software, provided that any modifications are also made open source under the same terms.
Commercial License: For proprietary use, internal corporate deployment without copyleft restrictions, or integration into closed-source workflows, a commercial license is available.

Please see LICENSE for details.

For commercial licensing inquiries, please contact: Taras Greben taras.greben@gmail.com.

📦 Third-Party Libraries

OpenCV.js: Computer Vision (Apache 2.0 License)
JSZip: File compression/decompression (MIT License)

See THIRD_PARTY_LICENSES.md for full text.

3.2 KiB Raw Blame History