Skip to content

Latest commit

 

History

History
44 lines (27 loc) · 2.18 KB

README.md

File metadata and controls

44 lines (27 loc) · 2.18 KB

Edge Detection and Gradient Image Generation

This C program performs edge detection and gradient image generation using pre-computed horizontal and vertical differential images. It implements edge detection by identifying points where the gradient magnitude exceeds a specified threshold, marking them as edges in the output image. The gradient image is computed from the horizontal and vertical differential images using a 3x3 or 5x5 filter.

Description

The program consists of two main functions:

  1. computeGradient: Computes the gradient image from the horizontal and vertical differential images. It calculates the magnitude of the gradient at each pixel using the Euclidean distance formula.

  2. detectEdges: Detects edge points in the gradient image based on a specified threshold. If the magnitude of the gradient at a pixel exceeds the threshold, it is marked as an edge point with a bright intensity (200). Otherwise, it is set to zero.

The program also includes utility functions to load images from files and perform memory allocation and deallocation.

Usage

  1. Compile the Program: Compile the program using a C compiler such as GCC.

    Example:

    gcc edge_detection.c -o edge_detection -lm
    
  2. Run the Program: Execute the compiled program.

    Example:

    ./edge_detection
    
  3. Input Images: Ensure that the input horizontal and vertical differential images are provided in raw format. Adjust the file paths in the code accordingly.

  4. Output Images: The program generates gradient images and edge images for both 3x3 and 5x5 filters. The output images are saved as raw files.

Sources and Dependencies

  • MTCNN for Face Detection: Uses pre-trained models from the Facenet PyTorch repository.
  • OpenCV for Image Processing: Utilizes OpenCV library for image loading and manipulation.
  • Scikit-learn for Cosine Similarity Calculation: Depends on Scikit-learn library for cosine similarity calculations.

Notes

  • Overflow and underflow are handled by normalizing the filtered results into the range of a byte.
  • Threshold values for edge detection can be adjusted manually through trial-and-error to achieve desired edge results.