Implementation of Snake Algorithmm for finding contours of Hand

I am following a research paper in which contour of hand is determined as shown in Figure 4.

In this paper, snakes were used to perform segmentation. A snake can be defined as an energy-minimizing spline guided by external restraining forces and influenced by image forces, which tends to localize it in features such as lines and edges. It is therefore an active contour that evolves dynamically relevant to the contours of the image. The snake is represented as a parametric curve by r(s) = (x(s),y(s)), with s(0, 1). Its functional energy can be expressed as:

Where Eint represents the internal energy of the snake, Eimg image forces and Eres external restraining forces.

Can you help in coding for this?

This is the image whose contours I need to find.

[Walter Roberson]

Please be more specific about "Can you help in coding for this?"

http://www.mathworks.com/matlabcentral/answers/6200-tutorial-how-to-ask-a-question-on-answers-and-get-a-fast-answer

[Explorer]

I need to make code for finding contour of hand as shown in figure 4 using snake algorithm.

How should I proceed?

[Walter Roberson]

Asking how to proceed is an algorithms or theory question, not a MATLAB question.

[Amelia Kartika]

How this coding for procced

Answer 1

[Image Analyst]

See my activecontour demo, attached. It will find the hand.

[Explorer]

facing following error

Undefined variable "images" or class "images.activecontour.internal.ActiveContourSpeedEdgeBased". Error in activecontour (line 222) speed = images.activecontour.internal.ActiveContourSpeedEdgeBased ... Error in activecontour_demo (line 68) bw = activecontour(grayImage, mask, 400, 'edge');

[Image Analyst]

Type

which activecontour
ver

You need to have a recent version of the Image Processing Toolbox. If you don't have one you should upgrade, unless you want to reinvent the wheel and spend a long time to write complicated snake code all by yourself.

[Explorer]

I have Image Processing Toolbox, Version 8.1 (R2012b)

Do you still think I need to upgrade?

[Image Analyst]

I have a good memory but I don't remember what versions all the various functions they've added first appeared. The key thing is the which command. Did you run it like I suggested? That's your answer. If it says you don't have it, then you need to upgrade.

[Explorer]

Okay, I have upgraded it. But still not getting expected output.

Here it is:

[Image Analyst]

Well you need to adapt it. Did you change the threshold or do something to get a binary image? Yours is just all white so it looks like you didn't do anything to get the proper starting binary image. For example, did you try this:

mask = grayImage < 100; % Use for hand

And you should also reduce the resolution to speed it up. It takes several minutes now and there is no need. You could get much faster times by reducing your resolution by a lot.

[Explorer]

I have done some modifications in code in order to get contour of desired image.

Problem:

1. Not getting perfect contour

2. I need hand contour image as binary image like this

________________________________________________________________________

clc; % Clear the command window.

close all; % Close all figures (except those of imtool.)

imtool close all; % Close all imtool figures if you have the Image Processing Toolbox.

clear; % Erase all existing variables. Or clearvars if you want.

workspace; % Make sure the workspace panel is showing.

format long g;

format compact;

fontSize = 15;

image=imread('C:\Users\Explorer\Documents\MATLAB\ASL_signs\A.jpg');

rimage=imresize(image, 0.20);

grayImage = rgb2gray(rimage);

% Get the dimensions of the image.

% numberOfColorBands should be = 1.

[rows, columns, numberOfColorBands] = size(grayImage);

if numberOfColorBands > 1

% It's not really gray scale like we expected - it's color.

% Convert it to gray scale by taking only the green channel.

grayImage = grayImage(:, :, 2); % Take green channel.

end

% Display the original gray scale image.

subplot(2, 2, 1);

imshow(grayImage, []);

axis on;

title('Original Grayscale Image', 'FontSize', fontSize);

% Enlarge figure to full screen.

set(gcf, 'Units', 'Normalized', 'OuterPosition', [0 0 1 1]);

% Give a name to the title bar.

set(gcf, 'Name', 'Demo by ImageAnalyst', 'NumberTitle', 'Off')

drawnow;

% Let's compute and display the histogram.

[pixelCount, grayLevels] = imhist(grayImage);

subplot(2, 2, 2);

bar(grayLevels, pixelCount);

grid on;

title('Histogram of original image', 'FontSize', fontSize);

xlim([0 grayLevels(end)]); % Scale x axis manually.

% % Construct an initial mask that is the convex hull of everything.

% subplot(2, 2, 2);

% % mask = grayImage < 175; % Use for eight.tif

% mask = grayImage > 150; % Use for coins.png

% mask = bwconvhull(mask, 'Union');

%imshow(mask);

% axis on;

% title('Initial mask', 'FontSize', fontSize);

% drawnow;

% Construct an initial mask that is the convex hull of everything.

subplot(2, 2, 2);

mask = sd_ycbcr(rimage) ; % skin detection function

imshow(mask);

axis on;

title('Initial mask', 'FontSize', fontSize);

drawnow;

% Now find the improved outer boundary using active contours.

bw = activecontour(grayImage, mask, 400, 'edge');

subplot(2, 2, 3);

imshow(bw);

axis on;

title('Outer Boundary Mask', 'FontSize', fontSize);

drawnow;

% Display the original gray scale image in the lower left

% So we can display boundaries over it.

subplot(2, 2, 4);

imshow(grayImage, []);

axis on;

hold on;

% Display the initial contour on the original image in blue.

% Display the final contour on the original image in red.

contour(mask,[0 0],'b', 'LineWidth', 4);

contour(bw,[0 0],'r', 'LineWidth', 4);

title('Image with initial and final contours overlaid', 'FontSize', fontSize);

________________________________________________________________________

Output:

[Image Analyst]

First of all, start with a good image. Notice how the example had a nice uniform background while yours has pixels both brighter, darker, and the same as the hand. Maybe just try an edge detection and fill it in, that is if you can't get a uniform background where you can simply threshold. For example, maybe you can do a color segmentation to get just the hand. Compute hsv, then look for pixels with high saturation. The sleeve and background will have low saturation because they are neutral colored. If you can threshold, just threshold the saturation image and call bwboundaries().

[Explorer]

Thanks for commenting.

I have written few more lines in code to get figure like figure 'b' shown above.

But how can I save figure 2 (shown below) in a variable to do further operations?

Code is also attached.

[Image Analyst]

You can use export_fig to save the figure with the axes, tickmarks, title, etc. Or you can use imwrite() to save just the image alone to a disk file. Then you can use imread() to read it back into a variable, though I don't know why you'd want figure stuff (title, etc.) as part of an image variable in your program.

[Image Analyst]

Well you saved "blank" - an image of all zeros. So what were you expecting?

[Explorer]

Sorry, last comment was deleted by me mistakenly.

[Explorer]

Please provide demo code for this

First of all, start with a good image. Notice how the example had a nice uniform background while yours has pixels both brighter, darker, and the same as the hand. Maybe just try an edge detection and fill it in, that is if you can't get a uniform background where you can simply threshold. For example, maybe you can do a color segmentation to get just the hand. Compute hsv, then look for pixels with high saturation. The sleeve and background will have low saturation because they are neutral colored. If you can threshold, just threshold the saturation image and call bwboundaries().

[Image Analyst]

HSV demo is here: http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862.

You can also try imgradient() which probably has demo code in the help.

[Amy Phan]

I tried this code but I'm not receiving any outline for my image. This is the outcome:

The image I'm using is an MRI image and I'm trying to use the snake method to outline the heart but I'm having issues with the thresholding. Any suggestions?

[Nhat Nguyen Thanh Minh]

try other threshold values

Answer 2

[snehal jaipurkar]

Sir how to save the segmented image after applying snake model in Matlab?????

[Image Analyst]

Use imwrite() or save().

Answer 3

[snehal jaipurkar]

And also I want to create a mask of the segmented image by putting the segmented part with intensity 1 and the rest of the part as zero. So how can I do that... Please please tell as soon as possible. Thanks alot in advance.

[Image Analyst]

My code does that. It's the "bw" in this line of code:

bw = activecontour(grayImage, mask, 400, 'edge');