Getting the edge pixels from the image

Question

0.00/5 (No votes)

See more:

Hi,
I would like to know any algorithm which will allow me to write the edge pixels(edges of the object in the image) from an image into a text file. I want this algorithm for my stego application.

ie, i want to extract the objects edge from the background.

try this link: i dont knw how to tel the exact requirement
http://www.cogs.susx.ac.uk/users/davidy/teachvision/vision3_include/fig17.gif

Posted 12-Sep-11 0:14am

Thiagarajan Duraisamy

Updated 13-Sep-11 5:07am

v3

Add a Solution

Comments

Herman<T>.Instance 12-Sep-11 6:28am

for c or c# (looking a your tags)?

Thiagarajan Duraisamy 13-Sep-11 11:15am

i need the algorithm alone or to say the idea, i want to work on myself to achieve the solution, wil try to code myself. so ideas will be gr8.

C_Johnson 14-Sep-11 17:26pm

Mainly for others if you did want to approach writing the algorithm yourself then you will need to read J Cannys paper A computational approach to edge detection available here as it is one of the most effective:

http://www.limsi.fr/Individu/vezien/PAPIERS_ACS/canny1986.pdf

Effectively what the algorithm does is examine each pixel with a set of 3x3 masks (i.e 1 pixel and the surrounding pixels) Each area is examined for edge cues at 0,90 and 45 degree angles. Effectively looking for a change in gradient. If there is a significant change in gradient (this is the thresholds you adjust) then an edge pixels is recorded as a 1 in a blank image. What you end up with is a binary map of edges. The code is not as easy as you would expect to write. On top of this you need to include a gaussian filter to remove noise and histogram analysis to further account for noise variation in pixel intensity's. As I said using existing libraries such as EMGU and openCV are far more efficient else you will be writing code again. Alternatively there is the AForge libraries that are simpler to work with but are not as advanced or powerful. See this article

http://www.codeproject.com/KB/GDI-plus/Image_Processing_Lab.aspx

Take Care

Thiagarajan Duraisamy 16-Sep-11 9:12am

k thanks, tc

3 solutions

Solution 3

you can use CImage object

#include <atlimage.h>

CImage TempObj;
TemObj.Load("Image File name");

//you can get Dimesion from
TempObj.GetHeight();
TempObj.GetWidth();

//To GetPixel Data
TempObj.GetPixel(X,Y);

Posted 13-Sep-11 1:54am

abhishek.biradar

Comments

abhishek.biradar 13-Sep-11 7:57am

Sorry its in c++

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

C_Johnson · Accepted Answer · 2011-09-12T09:25:00

Hi,

In c# your looking at something like bellow, personally I would write a Xaml file as then you could use tags for Red Green Blue. Looping through the image is the only option do not write as you get the pixel data as this will slow your program down since writing to your disc is relatively slow.

You could always use 4 lists one for top, bottom, left and right if you wish to keep the points separate. Alternatively positions [0 to (myImage.Width-1)] will be the top. [(myImage.Width-1) to (myImage.Width-1)+(myImage.Width-1)] will be the bottom and so on...

C#

Bitmap my_new_image = new Bitmap(original_image.Width,  original_image.Height);

List<color> my_colour_list = new List<colour>(); //somewhere to store the info as writing the information will take a long time programmatically speaking as your accessing hard drive 

//Top
for (int x = 0; x < myImage.Width; x++)
{
   my_colour_list.Add(output_data.GetPixel(x, 0))    
}

//bottom
for (int x = 0; x < myImage.Width; x++)
{
   my_colour_list.Add(output_data.GetPixel(x, myImage.Height))
}

//Left
for (int y = 0; y < myImage.Height; y++)
{
   my_colour_list.Add(output_data.GetPixel(0, y))
}

//Right
for (int y = 0; y < myImage.Height; y++)
{
   my_colour_list.Add(output_data.GetPixel(myImage.Width, y))
}

//Write the info to a text file (You will have Red, Green, Blue for colour image)
//It's up to you how you make the format

System.IO.StreamWriter file_to_write = new System.IO.StreamWriter("c:\\test.txt");

foreach(Color var in my_colour_list)
{
    //To save R/G/B and Alpha
    file_to_write.WriteLine(var.toString());
    //or just R G B
    file_to_write.WriteLine(var.R.toString());
    file_to_write.WriteLine(var.R.toString());
    file_to_write.WriteLine(var.R.toString());
}</colour></color>

Hope this helps,
Cheers,

Chris

lukeer · Accepted Answer · 2011-09-12T01:24:00

Solution 1

The edge pixels of any image should be

(0, 0)
(image.width - 1, 0)
(0, image.height - 1)
(image.width - 1, image.height - 1)

My fault, this gives corners, not edges.

Posted 12-Sep-11 1:24am

lukeer

Updated 12-Sep-11 21:24pm

v2