I am trying to extract features from an array representation of an image in MATLAB.
The features have a shape of a circle (ring) and a sector. This is shown in the image below. I have spent quite some time looking for a built-in function which does this. I have managed to do the ring extraction using an ugly looking loop but no idea where to start on the sector part. Any ideas how to implement this or even better a built-in function in MATLAB would be very helpful.
That's pretty easy, with no for loops needed, see for example in case your image is im:
[x y]=meshgrid(1:size(im,1));
f =#(x0,y0,r_max,r_min,theta1,theta2) ...
(x-x0).^2+(y-y0).^2<=r_max^2 & ...
(x-x0).^2+(y-y0).^2>=r_min^2 & ...
atan2(y-y0,x-x0)>=theta1 & ...
atan2(y-y0,x-x0)<=theta2;
f is a one liner anonymous function that accepts all needed parameters and gives a mask of the sector needed. For a ring you can set theta to be -pi to pi, or just delete the atan part from f. For example
r_max=40;
r_min=10;
x0=round(size(im,1)/2); %image center
y0=round(size(im,1)/2); %image center
theta1=deg2rad(10);
theta2=deg2rad(70);
imagesc(f(x0,y0,r_max,r_min,theta1,theta2))
set(gca,'YDir','normal')
axis square
Related
I am trying using MATLAB activecontour code to segment the region. the example was used grayscale image while i am using binary image. it turn out ok when i run the code by calling the binary image. however, when i combined the code, nothing happen. it skips the iteration part, and generate the sama binary image. for your reference, below is my code.
%% snake
figure
x = imread('1.jpg');
threshold = 160;
I = rgb2gray(x);
I = Igray>threshold;
imshow (I);
I = imresize(I,.5);
imshow(I)
title('Original Image')
mask = zeros(size(I));
mask(25:end-25,25:end-25) = 1;
imshow(mask)
title('Initial Contour Location')
bw = activecontour(I,mask,1300);
imshow(bw)
title('Segmented Image, 300 Iterations')
no process happen starting from snake's code. it eventually only generate binary image. I hope someone could try run this and help me to find my mistake. Thank you in advance
Matlab's activecontour function uses the Chan–Vese (active contours without edges) method by default, like Cris said. The implementation "uses the Sparse-Field level-set method, similar to the method described in [3]", citing Whitaker, "A level-set approach to 3d reconstruction from range data". (Besides Chan–Vese, activecontour has an optional method arg that can be set to 'edge' to use an alternative "edge-based model" based on the (older) geodesic active contours method of Caselles, Kimmel, and Sapiro.)
The Chan–Vese method segments a grayscale image by looking for a binary image equal to "c1" inside the contour and "c2" outside the contour that both has a smooth contour and is a good approximation to the original image. The method optimizes c1, c2, and the shape of the contour, beginning from some initial contour and evolving it by an iterative process.
If you'll excuse a self-citation, you can find an article, open source C code, and online demo about Chan–Vese on the IPOL journal at http://www.ipol.im/pub/art/2012/g-cv/, which you may find helpful.
So why is it not working in your case? Some thoughts:
In your use, since the input image is already binary, it is clearly tempting for the method to simply set c1=0, c2=1, and the contour to the edges of the input, so "nothing happens". Try setting the optional 'SmoothFactor' arg (possibly to a large value) to force the method to look for a smoother contour.
It's conceivably a datatype problem, since image I is passed as a logical array to activecontour, but normally the function takes a numeric array. Try casting I to a double array before passing.
I'm trying to find a starting point, but I can't seem to find the right answer. I'd be very grateful for some guidance. I also don't know the proper terminology, hence the title.
I took an image of a bag with a black background behind it.
And I want to extract the bag, similar to this.
And if possible, find the center, like this.
Essentially, I want to be able to extract the blob of pixels and then find the center point.
I know these are two separate questions, but I figured if someone can do the latter, then they can do the first. I am using MATLAB, but would like to write my own code and not use their image processing functions, like edge(). What methods/algorithms can I use? Any papers/links would be nice (:
Well, assuming that your image only consists of a black background and a bag inside it, a very common way to perform what you're asking is to threshold the image, then find the centroid of all of the white pixels.
I did a Google search and the closest thing that I can think of that matches what you want looks like this:
http://ak.picdn.net/shutterstock/videos/3455555/preview/stock-footage-single-blank-gray-shopping-bag-loop-rotate-on-black-background.jpg
This image is RGB for some reason, even though it's grayscale so we're going to convert this to grayscale. I'm assuming you can't use any built-in MATLAB functions and so rgb2gray is out. You can still implement it yourself though as rgb2gray implements the SMPTE Rec. 709 standard.
Once we read in the image, you can threshold the image and then find the centroid of all of the white pixels. That can be done using find to determine the non-zero row and column locations and then you'd just find the mean of both of them separately. Once we do that, we can show the image and plot a red circle where the centroid is located. As such:
im = imread('http://ak.picdn.net/shutterstock/videos/3455555/preview/stock-footage-single-blank-gray-shopping-bag-loop-rotate-on-black-background.jpg');
%// Convert colour image to grayscale
im = double(im);
im = 0.299*im(:,:,1) + 0.587*im(:,:,2) + 0.114*im(:,:,3);
im = uint8(im);
thresh = 30; %// Choose threshold here
%// Threshold image
im_thresh = im > thresh;
%// Find non-zero locations
[rows,cols] = find(im_thresh);
%// Find the centroid
mean_row = mean(rows);
mean_col = mean(cols);
%// Show the image and the centroid
imshow(im); hold on;
plot(mean_col, mean_row, 'r.', 'MarkerSize', 18);
When I run the above code, this is what we get:
Not bad! Now your next concern is the case of handling multiple objects. As you have intelligently determined, this code only detects one object. For the case of multiple objects, we're going to have to do something different. What you need to do is identify all of the objects in the image by an ID. This means that we need to create a matrix of IDs where each pixel in this matrix denotes which object the object belongs to. After, we iterate through each object ID and find each centroid. This is performed by creating a mask for each ID, finding the centroid of that mask and saving this result. This is what is known as finding the connected components.
regionprops is the most common way to do this in MATLAB, but as you want to implement this yourself, I will defer you to my post I wrote a while ago on how to find the connected components of a binary image:
How to find all connected components in a binary image in Matlab?
Mind you, that algorithm is not the most efficient one so it may take a few seconds, but I'm sure you don't mind the wait :) So let's deal with the case of multiple objects now. I also found this image on Google:
We'd threshold the image as normal, then what's going to be different is performing a connected components analysis, then we iterate through each label and find the centroid. However, an additional constraint that I'm going to enforce is that we're going to check the area of each object found in the connected components result. If it's less than some number, this means that the object is probably attributed to quantization noise and we should skip this result.
Therefore, assuming that you took the code in the above linked post and placed it into a function called conncomptest which has the following prototype:
B = conncomptest(A);
So, take the code in the referenced post, and place it into a function called conncomptest.m with a function header such that:
function B = conncomptest(A)
where A is the input binary image and B is the matrix of IDs, you would do something like this:
im = imread('http://cdn.c.photoshelter.com/img-get2/I0000dqEHPhmGs.w/fit=1000x750/84483552.jpg');
im = double(im);
im = 0.299*im(:,:,1) + 0.587*im(:,:,2) + 0.114*im(:,:,3);
im = uint8(im);
thresh = 30; %// Choose threshold here
%// Threshold image
im_thresh = im > thresh;
%// Perform connected components analysis
labels = conncomptest(im_thresh);
%// Find the total number of objects in the image
num_labels = max(labels(:));
%// Find centroids of each object and show the image
figure;
imshow(im);
hold on;
for idx = 1 : num_labels
%// Find the ith object mask
mask = labels == idx;
%// Find the area
arr = sum(mask(:));
%// If area is less than a threshold
%// don't process this object
if arr < 50
continue;
end
%// Else, find the centroid normally
%// Find non-zero locations
[rows,cols] = find(mask);
%// Find the centroid
mean_row = mean(rows);
mean_col = mean(cols);
%// Show the image and the centroid
plot(mean_col, mean_row, 'r.', 'MarkerSize', 18);
end
We get:
I have no intention of detracting from Ray's (#rayryeng) excellent advice and, as usual, beautifully crafted, reasoned and illustrated answer, however I note that you are interested in solutions other than Matlab and actually want to develop your own code, so I though I would provide some additional options for you.
You could look to the excellent ImageMagick, which is installed in most Linux distros and available for OS X, other good operating systems and Windows. It includes a "Connected Components" method and if you apply it to this image:
like this at the command-line:
convert bags.png -threshold 20% \
-define connected-components:verbose=true \
-define connected-components:area-threshold=600 \
-connected-components 8 -auto-level output.png
The output will be:
Objects (id: bounding-box centroid area mean-color):
2: 630x473+0+0 309.0,252.9 195140 srgb(0,0,0)
1: 248x220+0+0 131.8,105.5 40559 srgb(249,249,249)
7: 299x231+328+186 507.5,304.8 36620 srgb(254,254,254)
3: 140x171+403+0 458.0,80.2 13671 srgb(253,253,253)
12: 125x150+206+323 259.8,382.4 10940 srgb(253,253,253)
8: 40x50+339+221 357.0,248.0 1060 srgb(0,0,0)
which shows 6 objects, one per line, and gives the bounding boxes, centroids and mean-colour of each. So, the 3rd line means a box 299 pixels wide by 231 pixels tall, with its top-left corner at 328 across from the top-left of the image and 186 pixels down from the top-left corner.
If I draw in the bounding boxes, you can see them here:
The centroids are also listed for you.
The outout image from the command above is like this, showing each shape shaded in a different shade of grey. Note that the darkest one has come up black so is VERY hard to see - nearly impossible :-)
If, as you say, you wish to look at writing your own connected component code, you could look at my code in another answer of mine... here
Anyway, I hope this helps and you see it just as an addition to the excellent answer Ray has already provided.
I am using matlab to process data to get a radius distribution function. Now I get data(an array) of different distances of other atoms to 1 specific atom.
I used the "hist" command (hist(radius1,400)) and get a histogram:
But what I want is a curve, like this:
http://upload.wikimedia.org/wikipedia/commons/3/31/Lennard-Jones_Radial_Distribution_Function.svg
I tried some fit command, but it would give me a normal-distribution-like curve, which is not what I want. actually no fit is fine, I only want a curve to show its varying.
The raw data was a 4000*1 array of radius, is there any other way to get a curve of the top of each bar of the histogram?
Thanks so much.
Instead of automatically plotting a histogram using hist, you can get it to output the values:
[x, c] = hist(radius1,400);
x is the data in each bin, c the centre of each bin, so this replicates a histogram and then overplots a line on it (which will just connect the top of each bar so it may not look as smooth as you hoped):
bar(c,x);
hold on
plot(c,x,'r');
It is possible to use fit with an anonymous function as a custom model, but that may be overkill in this situation.
I got a code from the book Feature Extraction & Image Processing.
As I am a total beginner in Matlab, I don't know how to run these codes to see results.
Are they complete?
First one : Hough Transform for Lines
%Polar Hough Transform for Lines
function HTPLine(inputimage)
%image size
[rows,columns]=size(inputimage);
%accumulator
rmax=round(sqrt(rows^2+columns^2));
acc=zeros(rmax,180);
%image
for x=1:columns
for y=1:rows
if(inputimage(y,x)==0)
for m=1:180
r=round(x*cos((m*pi)/180)+y*sin(m*pi)/180);
if(r0) acc(r,m)=acc(r,m)+1; end
end
end
end
end
Second one : Hough Transform for Circles
%Hough Transform for Circles
function HTCircle(inputimage,r)
%image size
[rows,columns]=size(inputimage);
%accumulator
acc=zeros(rows,columns);
%image
for x=1:columns
for y=1:rows
if(inputimage(y,x)==0)
for ang=0:360
t=(ang*pi)/180;
x0=round(x-r*cos(t));
y0=round(y-r*sin(t));
if(x00 & y00)
acc(y0,x0)=acc(y0,x0)+1;
end
end
end
end
end
Third one : Hough Transform for Elipses
%Hough Transform for Ellipses
function HTEllipse(inputimage,a,b)
%image size
[rows,columns]=size(inputimage);
%accumulator
acc=zeros(rows,columns);
%image
for x=1:columns
for y=1:rows
if(inputimage(y,x)==0)
for ang=0:360
t=(ang*pi)/180;
x0=round(x-a*cos(t));
y0=round(y-b*sin(t));
if(x00 & y0< rows & y0>0)
acc(y0,x0)=acc(y0,x0)+1;
end
end
end
end
end
I have images (png) that I need to run these programs with.
But I cannot seem to run it.
I create new script, paste the code, save it and in main window I run the function name sending path to the image as a parameter. It does nothing, no message or so.
Your functions don't return any value, that means that you have to add a return argument to those functions (in case of the hough trafo, you'd like to return the accumulator array acc) and as described in the manual (http://www.mathworks.de/de/help/matlab/ref/function.html), change the functions headers to:
function acc = HTPLine(inputimage)
and then also call it like this from the commandline (or from another script):
IMG = imread('some_image.jpg');
% e.g. convert to grayscale:
IMG = rgb2gray(IMG);
acc = HTPLine(IMG);
you then still need to find the maximum (or several maxima, depending on how many lines you'd like to fit) in the accumulator and display the fitted line (ellipse/circle) on your own via plot in a figure...
EDIT: Looking at your code, I don't know what the variable r0 should be!? It will def. give you an errormsesage as it's not defined anywhere.
And why do you the if(inputimage(y,x)==0) in your code? The pixel at the current point doesn't have to be black in ordner to calculate the hough transformation for that single pixel (just think of a grey-image with a white line in it -> your code wouldn't do anything as the image wouldn't contain any black pixels). So that's what you need to rethink about.
If you want to, you could also use the built-in hough transformation of MATLAB: http://www.mathworks.de/de/help/images/ref/hough.html // http://www.mathworks.de/de/help/images/ref/houghpeaks.html
please help with Matlab beginner challenge
i need to create an image with few geometrical objects (circles,
ellipses) and then to apply some projective transforms
my problem is that i cant understand how to actually "draw" on image
image is AFAIU generally defined as [X;Y;3] matrix,
functions as SCIRCLE1 can compute/return collection of points
representing circle, but the problem is that points are not discrete ,
coordinates are real numbers and not pixels
how can i recompute the scircle output to be valid in image
coordinates system ? i.e. how can i "pixelize" it?
thanks for your attention, i really missing some basic concept and
will appreciate your help
John
well, below is an answer i received on Matlab newsgroups
BOTTOM LINE-no built-in way in Matlab
======================================
'getframe' can be used to merge axes even though it is more commonly used to create movie frames.
MATLAB is really weak in this area. There are some primitive
functions for drawing into the overlay (such as rectangle() if you
want to draw a circle, and line() if you want to draw a line) but no
real way that I know of to draw right into the underlying image. So
you have to use "tricks" such as getframe and then apply logical
operations. And you have to be careful with that since I think when
it gives you the rasterized version of the overlay it might be the
size of the image on the screen, not the true original matrix size of
the underlying image (I'd have to recheck this).
full thread here : http://www.mathworks.com.au/matlabcentral/newsreader/view_thread/261232
I found this example which give you a easy way to put simple geometrical object onto pictures.
Read the input image.
I = imread('cameraman.tif');
Define the rectangle dimensions as [x y width height].
rectangle = int32([10 10 30 30]);
Draw the rectangle and display the result.
J = step(shapeInserter, I, rectangle);
imshow(J);
see this link
by the way..
I didn't get you point about points not being discrete and images being a matrix. The way I see it. It much the same. you could try to explain it more in depth ?
The insertShape function in the Computer Vision System Toolbox is what you need. It lets you draw rectangles, circles, and polygons into the image.
There is also insertText, insertMarker, and insertObjectAnnotation.