Edge Detection Project

Overview

This project implements edge detection algorithms in C#, featuring Sobel, Prewitt, and Roberts operators. Edge detection is a fundamental image processing technique that identifies boundaries of objects within images by detecting discontinuities in brightness.

Supported Operators

• Sobel Operator: Emphasizes edges using two 3x3 kernels for horizontal and vertical edge detection, with stronger weight on central pixels
• Prewitt Operator: Similar to Sobel but with uniform weights, making it more sensitive to noise but potentially better at detecting diagonal edges
• Roberts Operator: Uses 2x2 kernels for fast computation, best suited for sharp edges and high-contrast images

Requirements

• .NET 8.0 or higher
  • Download .NET Framework 8.0 SDK
• Windows OS (Required for System.Drawing dependencies)

Installation & Setup

1. Clone the repository:

   git clone https://github.com/steefpls/EdgeDetection.git
   cd EdgeDetection

2. Build the application:

   dotnet build

3. Run the tests:

   dotnet test

Usage

1. Prepare your image:

   • Place your input image in this folder (EdgeDetection.ConsoleApp\bin\Debug\net8.0)
   • Supported formats: JPG, PNG
   • Images are automatically converted to grayscale during processing

2. Run the application:

   dotnet run --project EdgeDetection.ConsoleApp <input_image> <output_image>

   Example:

   dotnet run --project EdgeDetection.ConsoleApp input.jpg output.png

3. Select operator:

   • When prompted, choose: 1: Sobel operator 2: Prewitt operator 3: Roberts operator

Example Output

Original Image	Sobel Output	Prewitt Output	Roberts Output

Architecture

The project follows SOLID principles and clean architecture:

• Core Components:

  • GrayscaleImage: Handles image data in grayscale format
  • BaseOperator: Abstract class implementing common edge detection logic
  • IEdgeDetector: Interface for edge detection operators
  • IImageProcessor: Interface for image processing operations

• Operators:

  • Each operator (Sobel, Prewitt, Roberts) extends BaseOperator
  • Implements specific kernel matrices for edge detection
  • Factory pattern used for operator instantiation

• Processing Pipeline:

  1. Image loading and grayscale conversion
  2. Operator selection and initialization
  3. Edge detection processing
  4. Result saving

Testing

The project includes comprehensive unit tests covering:

• Operator functionality
• Image processing pipeline
• Edge cases and error handling
• Input validation

Test coverage includes:

• All edge detection operators
• Image processing workflows
• Factory class functionality
• Grayscale image operations

Known Limitations

1. Image Size:

   • Very large images may require significant memory
   • Border pixels are not processed

2. Platform:

   • Windows-only due to System.Drawing dependencies
   • No Linux/macOS support currently

3. Performance:

   • Single-threaded processing
   • No GPU acceleration

Troubleshooting

Common issues and solutions:

1. FileNotFoundException:

   • Ensure input file exists in the correct directory
   • Check file permissions

2. OutOfMemoryException:

   • Try processing a smaller image
   • Ensure sufficient system memory

UML Diagram