图像处理代写 image processing代写 python代写

image processing python

Implementation (50 %)

图像处理代写 1 DoG (20 %) Use two Gaussian filters with suitable kernel sizes to to extract the license plate and remove most of the background for ”lp.jpg”. 1.1

1 DoG (20 %) 图像处理代写

Use two Gaussian filters with suitable kernel sizes to to extract the license plate and remove most of the background for ”lp.jpg”.

1.1

Apply canny operator on the image to extract the edges. Try to remove the noise by preprocess the image using a gaussian filter. Try to keep the license plate letters and numbers while removing other parts of the backgrounds by adjusting two thresholds in Canny operator.

2

Which of the previous methods do you think is more useful to remove the background when we do have information about the scale our object? Why?

3 Template Matching (30 %) 图像处理代写

Write a program that takes a template and an image. Then perform template matching using normalized cross correlation. Visualize heat map that shows the probability of having circle (using template ”circle.bmp”) in ”messi.jpg”. Where is the peak in the heat map and what does it show? You can use the edge of the image using Canny operator to get a better result.

3.1

Compare your function with ”matchTemplate” function in OpenCV. You should normalize the output matrix of ”matchTemplate”. Set the fourth argument of method ”matchTemplate” to ”CV TM CCOEFF NORMED”.

3.2

Propose a solution to find circles at different scales. Now implement your solution and find ”circle.bmp” on the resized ”messi.jpg” with scale factor ”2” to show your method works.

更多代写：代修CS网课多伦多  toefl代考  英国代做政治学作业  分析类论文代写  教育学作业写作 代写英国论文写作 

合作平台：essay代写 论文代写 写手招聘 英国留学生代写