Calculate the cone angle of a spray image

Please find the image bellow - the angle is between the boundary lines in red shown, while the reference point is the center point in the line shown as reference. The reference point is at fixed distance known value. The dark fish shaped is the spray shape after impingement- for which we need to measure the angle. Scattered dots are particle in liquid form-

To tackle I thought to get following steps: read the image Subtracting image with background image to get clear spray visualization convert grey-scale to binary setting threshold for separating the dark and light area- for boundary Drawing the boundary lines as spray after collision and calculating the angle between lines

[Joseph Cheng]

It sounds like you're on the right track. What's the question?

[Image Analyst]

Attach the original, un-annotated image and your script if you need help on it.

[sanjeet]

My actual image to be processed is

My matlab code:

% Calculate the cone angle of a spray image
clear, clc, close all
  Image = imread(['44_M_b.jpg']);
  [a1 a2 a3]=size(Image);
  x0=215; %if picture higher than increase number  215
  y0=255; %if picture to the left than increase number  255
  r=170;  %170
   Image2=imadjust(rgb2gray(Image));
  figure(1); imshow(Image2)
  hold on
  BW = im2bw(Image2);
  figure(2); imshow(BW)
  BW1 = edge(BW,'sobel');
  figure(3); imshow(BW1);
  size(BW1)

I am getting the image as

My next step is to get scan pixel to get a line to represent the spray edge as shown in image earlier. They should start from intersection at reference point. I tried using bwtraceboundary and find but unable to create boundary along the edge inner. Please guide me to get the line along the desired spray edge.

Answer 1

[Image Analyst]

Alright sanjeet, I spent way more time than I should have to make this fancy for you (about half an hour) but attached below the image is the code.

It determined that the angle is 23.1 degrees. If you want the lines overlaid on the image, I'm sure you can figure out that simple part. As I mentioned in the comments you might want to figure out an algorithm to examine the widths plot and figure out where to fit your line over. The min is fairly straight forward, but the widest part is somewhat of a judgement call.

[sanjeet]

Thanks a lot

I got 24 degree as cone angle on increasing the threshold value to 65 from 40... which matches with that measure by protector using paint shop

Answer 2

[Dhruv Gandhi]

Hallo image analyst,

i want to calculate the area of the coloured image , see the attachment.

i try to convert all the blue lines into white and other then Blue in to Black. after That want to calculateArea of this Black region in the image. i Dont know How to do that. can ypu please Help.

Answer 3

[I Komang Gede Tryas Agameru Putra]

Dear Image Analyst,

Thank you for providing the code above, I tried to use your code to analyze the cone angle, but it turns out that the cone angle is not properly calculated since it shows zero values (please kindly see the attachment).

Please kindly help me to solve this issues.

Thank you in advanced.

[Image Analyst]

@I Komang Gede Tryas Agameru Putra the biggest blob is your background (touching the border) so after you get the binary image the first time, you'll have to call imclearborder

help imclearborder

 IMCLEARBORDER Suppress light structures connected to image border.
    IM2 = IMCLEARBORDER(IM) suppresses structures that are lighter than
    their surroundings and that are connected to the image border.  IM can
    be an intensity or binary image.  The output image, IM2, is intensity or
    binary, respectively.  The default connectivity is 8 for two dimensions,
    26 for three dimensions, and CONNDEF(NDIMS(BW),'maximal') for higher
    dimensions.
 
    For intensity images, IMCLEARBORDER tends to reduce the overall
    intensity level in addition to suppressing border structures.
 
    IM2 = IMCLEARBORDER(IM,CONN) specifies the desired connectivity.
    CONN may have the following scalar values:  
 
        4     two-dimensional four-connected neighborhood
        8     two-dimensional eight-connected neighborhood
        6     three-dimensional six-connected neighborhood
        18    three-dimensional 18-connected neighborhood
        26    three-dimensional 26-connected neighborhood
 
    Connectivity may be defined in a more general way for any dimension by
    using for CONN a 3-by-3-by- ... -by-3 matrix of 0s and 1s.  The 1-valued
    elements define neighborhood locations relative to the center element of
    CONN.  CONN must be symmetric about its center element.
 
    Class Support
    -------------
    IM can be a numeric or logical array of any dimension, and it must be
    nonsparse and real.  IM2 has the same class as IM.
 
    Note
    ----
    A pixel on the edge of the input image might not be considered to be
    a "border" pixel if a nondefault connectivity is specified.  For
    example, if CONN = [0 0 0; 1 1 1; 0 0 0], elements on the first and
    last row are not considered to be border pixels because, according to
    that connectivity definition, they are not connected to the region
    outside of the image.
 
    References: 
    -----------
    P. Soille, Morphological Image Analysis: Principles and Applications,
    Springer, 1999, pp. 164-165.
 
    Example 1
    ---------
    % This example shows how to clear the border of an intensity image:
 
        I = imread('rice.png');
        I2 = imclearborder(I);
        montage({I,I2})
 
    Example 2
    ---------
    % This example shows how to clear the border of a binary image:
 
        BW = imbinarize(imread('rice.png'));
        BW2 = imclearborder(BW);
        montage({I,I2})
 
    Example 3
    ---------
    % This example shows how to retain some borders while clearing:
  
        I = imread('rice.png');
        % Specify border, in order from: top, right, bottom, left.
        bordersToKeep = [0 1 0 1]; % keep right and left
        % Top and left
        Ipad = padarray(I, [bordersToKeep(1), bordersToKeep(4)], 'pre');
        % Bottom and right
        Ipad = padarray(Ipad, [bordersToKeep(3), bordersToKeep(2)], 'post');
        I2 = imclearborder(Ipad);
        % Unpad
        topLeft = [bordersToKeep(1) bordersToKeep(4)]+1;
        botRight = topLeft + [size(I,1), size(I,2)] - 1;
        I2 = I2(topLeft(1):botRight(1), topLeft(2):botRight(2));
        montage({I,I2})
 
    See also IMRECONSTRUCT.

    Documentation for imclearborder
       doc imclearborder

[I Komang Gede Tryas Agameru Putra]

Dear Image Analyst,

Thank you very much for your advice, cone angle is now shown but I have further question.

How can I know if the calculated value of spray angle is measured based on the spray origin? (marked by the red pen in attachment 1), and is there a way to fixed the spray origin into the specific value?

Because when I imported the image without spray, the code still calculated the cone angle is around 19 degree even though there were no spray in the images. (Please kindly refer to the attachment 2)

The image that will be processed is also attached.

Please kindly guide me to understand this matter.

Thank you again for your support.

[Image Analyst]

Looks like your binary image needs to be reversed. Start your own question and attach the original image and your script if you need more help.

[I Komang Gede Tryas Agameru Putra]

Dear Image Analyst,

I have posted the question in separate section. Thank you in advanced

[Image Analyst]

Thanks. I'll look at it later today.