Edge_and_Corner_Detection

Lab 2: Feature Detection

Introduction

This project detects features in image i.e edges using Canny edge detection and interest points using Harris interest point detection.

The code was tested on Matlab 2017a.

• Please refer to Report.pdf for detailed analysis.
• Please refer to lab.pdf for about the project.

Directory Structure

---code

|
|---MyCannyEdgeDetector.m
|---MyCompareOutput.m
|---MyDetectInterest.m
|---lenna.png
|---test.jpg

---lab.pdf

---README.md

---Report.pdf

MyCannyEdgeDetector(image, threshold)

It returns the output canny image

image is image matrix variable

threshold is 2x1 vector in format [low high]

e.g. imshow(MyCannyEdgeDetector(imread('lenna.png'),[0.0375 0.0938])); imshow(MyCannyEdgeDetector(imread('test.jpg'),[0.03 0.2]));

MyCompareOutput(image, threshold)

It returns the MSE and PSNT matrix and saves the comparison figure as 'comp_output.jpg'

image is image matrix variable

threshold is 2x1 vector in format [low high]

e.g. MyCompareOutput(imread('lenna.png'),[0.0375 0.0938]); MyCompareOutput(imread('test.jpg'),[0.03 0.2]);

MyDetectInterest(image, threshold)

It saves the detected interest points figure as 'harris_detect.jpg'

image is image matrix variable

if threshold is 2x1 vector in format [low high] it is used a threshold for canny edge detector and high is used as threshold for Harris Detector

if threshold is 3x1 vector in format [low high harris_threshold] then [low high] is used a threshold for canny edge detector and harris_threshold is used as threshold for Harris Detector

e.g. MyDetectInterest(imread('test.jpg'),[0.03 0.2]); MyDetectInterest(imread('test.jpg'),[0.03 0.2 0.01]);

Image Reference

www.cse.iitd.ernet.in/~pkalra/col783/canny.pdf

https://en.wikipedia.org/wiki/Lenna

Developed By

Naman Goyal ([email protected])