Problem 1: EDGE DETECTION
In this problem, please design several edge detection algorithms to satisfy the following requirements.
(a) sample1.jpg
(b) sample2.jpg
Figure 1: Images for edge detection.
(a) Given an image, sample1.jpg.
(1) Perform 1st order edge detection and output the edge maps as result1.jpg
(2) Perform 2nd order edge detection and output the edge maps as result2.jpg
(3) Perform Canny edge detection and output the edge maps as result3.jpg
(4) Apply an edge crispening method to the given image, and output the result as result4.jpg. Please also generate an edge map of result4.jpg as result5.jpg.
(5) Compare result1.jpg, result2.jpg, result3.jpg and result5.jpg. Provide some discussions and findings in the report.
Provide some discussions about those methods. For example, their pros and cons, how parameters affect the results, etc. Please mark the edge pixels with intensity 255 and backgrounds pixels with intensity 0 on your results.
(b) Please design an algorithm to obtain the edge map of sample2.jpg as best as you can. Describe the steps in detail and provide some discussions.
Problem 2: GEOMETRICAL MODIFICATION
In this problem, please design several geometrical modification algorithms to meet the following requirements. Your results don’t have to be exactly the same as sample images, just try to make the effects.
(a) Please design an algorithm to make sample3.jpg become sample4.jpg. Output the result as result6.jpg. Please describe your method and implementation details clearly. (hint: you may perform rotation, scaling, translation, etc.)
(b) Imagine that there is a black hole in the center absorbing sample5.jpg. Please design a method to make sample5.jpg look like sample6.jpg as much as possible and save the output as result7.jpg.
(a) sample3.jpg
(b) sample4.jpg
(c) sample5.jpg (d) sample6.jpg
Figure 2: Images for geometrical modification.
Appendix
Problem 1: EDGE DETECTION
sample1.jpg: 1200 × 1650 gray-scale sample2.jpg: 1200 × 1650 gray-scale
Problem 2: GEOMETRICAL MODIFICATION
sample3.jpg:
1024 × 1280
gray-scale
sample5.jpg:
1024 × 1280
gray-scale
