Digitize your scanned images
This project was born out of a practical need: digitizing multiple family photo albums.
Like many families, we had boxes full of old photo albums collecting dust – precious memories that deserved to be preserved digitally. Commercial scanning software was either too expensive, too complicated, or simply didn't do what we needed: scan multiple photos at once and automatically crop them individually.
So I built Image Digitizer – a simple, robust tool that:
- Scans a full page with multiple photos
- Automatically detects and crops each photo
- Saves them individually without quality loss
What started as a weekend project to solve my own problem turned into a fully-featured application that I'm now sharing with everyone who faces the same challenge.
If you're sitting on a pile of photo albums waiting to be digitized – this tool is for you.
- Automatic detection of all WIA 2.0 scanners
- Support for high resolutions (up to 1200 DPI)
- No external manufacturer popup – everything in the app
- Detects multiple photos on a single scan
- Automatic cropping without quality loss
- OpenCV-based image processing
How it works:
- Converts the scan to grayscale and applies noise reduction
- Uses threshold detection to separate photos from the white background
- Applies morphological operations to group connected regions
- Filters detected regions by size and color variance to identify actual photos
- Automatically crops each detected photo with a small margin
Important Notes:
- Darker Images: Very dark photos or photos with very low contrast may not be detected correctly. The algorithm works best with photos that have clear contrast against a white or light background.
- Fallback Behavior: If no photos are detected (e.g., very dark images or poor contrast), the entire scan will be saved as a single file instead of individual photos. This ensures you don't lose your scan even if automatic detection fails.
- Logging: All detection results (successful detections, failed detections, number of photos found) are logged to
%LOCALAPPDATA%\ImageDigitizer\log.txtfor troubleshooting.
- PNG (lossless) or JPEG export
- Automatic saving after scan
- Custom storage location (remembered between sessions)
- German / English
- Modern toggle switch for switching
- Dynamic timeout based on DPI (high-res scans get more time)
- Automatic retry on busy scanner
- Improved connection cleanup
The app offers a modern, intuitive user interface in Windows 11 style.
- Windows 10/11
- WIA 2.0 compatible scanner
- No additional software required - .NET Runtime is included
- Download
ImageDigitizer-v1.0-win-x64-selfcontained.zipfrom Releases - Extract the ZIP archive
- Run
ImageDigitizer.exe - Done! No additional installation needed.
Note: The self-contained version includes the .NET Runtime (~150 MB), so no separate installation is required. It works on any Windows 10/11 PC.
If you want to build the application yourself, you have two options:
- .NET 8.0 SDK
- Visual Studio 2022, Rider, or any IDE with .NET support
- Git (for cloning the repository)
This creates a standalone executable with all dependencies included:
# Clone the repository
git clone https://github.com/mrhymes26/image-digitizer.git
cd image-digitizer
# Build self-contained version
dotnet publish -c Release -r win-x64 --self-contained true -p:SelfContained=true -o "release-build-selfcontained"The output will be in release-build-selfcontained/ directory. This version includes the .NET Runtime and all dependencies (~150 MB).
This creates a smaller build that requires .NET 8.0 Runtime to be installed:
# Clone the repository
git clone https://github.com/mrhymes26/image-digitizer.git
cd image-digitizer
# Build standard version
dotnet publish -c Release -r win-x64 -o "release-build"The output will be in release-build/ directory. Users will need to have .NET 8.0 Runtime installed (~10 MB).
For development and testing:
# Clone the repository
git clone https://github.com/mrhymes26/image-digitizer.git
cd image-digitizer
# Restore dependencies
dotnet restore
# Build
dotnet build
# Run
dotnet rungit clone https://github.com/mrhymes26/image-digitizer.git
cd image-digitizer
dotnet build
dotnet run- Framework: WPF (.NET 8)
- Scanner API: WIA 2.0 (Windows Image Acquisition)
- Image Processing: OpenCVSharp 4.8
- Language: C# 12
ImageDigitizer/
├── App.xaml # Application Resources & Styles
├── MainWindow.xaml # Main UI
├── MainWindow.xaml.cs # UI Logic
├── Models/
│ ├── ScannerDevice.cs # Scanner Data Model
│ └── ScanSettings.cs # Scan Settings
├── Services/
│ ├── ScannerService.cs # Scanner Abstraction (STA-Thread handling)
│ ├── WiaInterop.cs # WIA COM Interop
│ └── ImageProcessor.cs # Image Processing & Detection
└── Utils/
├── AppSettings.cs # Persistent Settings
├── Localization.cs # DE/EN Translations
└── Logger.cs # Debug Logging
Settings:
%LOCALAPPDATA%\ImageDigitizer\settings.json
Contains: output folder, language, last scanner, DPI, format preferences
Logs:
%LOCALAPPDATA%\ImageDigitizer\log.txt
Contains: debug information, scan operations, error messages
Output Folder (Scanned Images):
- Default:
%USERPROFILE%\Pictures\ImageDigitizer - Can be changed in the app settings
- Each scan creates individual photo files (PNG or JPEG)
- Scan: Place multiple photos on the scanner and click "Scan"
- Detection: OpenCV algorithms automatically detect individual photos
- Processing: Each detected photo is cropped with a small margin
- Saving: Photos are saved individually as separate files (PNG or JPEG)
- Ready: All photos are saved to your output folder
Fallback: If no photos are detected (e.g., very dark images), the entire scan is saved as a single file to ensure you don't lose your work.
The app uses computer vision algorithms to identify photo boundaries:
- Grayscale Conversion: Converts the scan to grayscale for processing
- Noise Reduction: Applies filters to reduce scanner noise
- Threshold Detection: Separates photos from the white/light background
- Morphological Operations: Groups connected regions together
- Size & Color Filtering: Identifies actual photos by size and color variance
- Auto-Cropping: Crops each detected photo with a small margin
Best Results:
- Photos with clear contrast against white/light background
- Some spacing between photos
- Good lighting on the scanner
- Clean scanner glass
Limitations:
- Very dark images may not be detected correctly
- Low contrast photos (e.g., dark photos on dark background) may fail detection
- If detection fails: The entire scan is saved as a single file, so you don't lose your work
- Logging: Detection results are logged to
%LOCALAPPDATA%\ImageDigitizer\log.txt- check this file if photos aren't detected as expected
- WSD scanners are often limited to 300 DPI
- High-res scans may take several minutes – the app now has dynamic timeouts
- All WIA 2.0 compatible scanners
- Tested with Brother MFC-L8690CDW
See CHANGELOG.md for full version history.
- WIA 2.0 scanner support with automatic detection
- Automatic photo detection and cropping using OpenCV
- Lossless saving (PNG) or high-quality JPEG (95%)
- Dynamic scan timeout based on DPI
- Robust STA-Thread handling with retry logic
- Windows 11 modern UI design
- German/English localization
Contributions are welcome! Please create a Pull Request or open an Issue.
See CONTRIBUTING.md for details.
This project uses the following third-party libraries:
-
OpenCvSharp4 (v4.8.0.20230708) - Apache License 2.0
- .NET wrapper for OpenCV image processing
- GitHub: https://github.com/shimat/opencvsharp
-
System.Drawing.Common (v8.0.0) - MIT License
- Microsoft .NET library for image handling
-
WIA 2.0 - Windows API
- Built-in Windows Image Acquisition API (no additional license)
For full license information, see NOTICE file.
MIT License - see LICENSE
Developed by mrhymes26
If you find this project useful, consider supporting its development:
Note: This application runs entirely locally on your computer:
- No network connections (except for scanner communication)
- No data is sent to external servers
- All processing happens on your machine
- Scanned images are saved only to your local storage
As such, security concerns are minimal. The application only accesses:
- Your scanner (via WIA 2.0)
- Local file system (for saving images)
- Windows system APIs
Image Digitizer – Digitize your memories, one album at a time.