Canny Edge Detector Implementation and Comparative Analysis
- Login to Download
- 1 Credits
Resource Overview
(1) Implement a Canny edge detector function using MATLAB (or C) with Gaussian filtering capabilities; (2) Apply this custom function to detect edges in images corrupted by salt-and-pepper or Gaussian noise; (3) Utilize MATLAB's built-in Canny edge detection function on the same noisy images; (4) Perform comparative analysis between both approaches to evaluate performance differences.
Detailed Documentation
This project requires completing several key steps for edge detection implementation:
First, we need to create a custom Canny edge detector function using MATLAB programming language, which incorporates Gaussian filtering operations. The implementation should include gradient calculation using Sobel operators, non-maximum suppression, and double thresholding with hysteresis tracking to connect weak edge pixels.
Next, we will apply our custom Canny function to perform edge detection on test images containing either salt-and-pepper noise or Gaussian noise. The Gaussian filtering stage will help reduce noise interference before edge detection.
Additionally, we will use MATLAB's built-in edge detection function (edge(image, 'canny')) with the same parameter settings on identical noisy images to establish a baseline for comparison.
Finally, we will conduct a comprehensive comparison between the results obtained from our custom implementation and MATLAB's native function. This evaluation will include quantitative metrics like edge connectivity analysis and qualitative assessment of noise resilience, providing insights into the performance characteristics of both approaches.
- Login to Download
- 1 Credits