I want to find the edge lengths of the shape in a given image.

I have an edged image and would like to highlight the edges of triangles and also find the length of each side of each triangle. Display the highlighted triangles and the length of each side of triangle individually.

The edges of the triangles are not straight but have few uneven (curvy) lines as well. I want to highlight all of them making the edges into a straight line. Also as there are different shapes other than triangles, I have to first detect it as a triangle (maybe wrt to angle between two lines) and then find the length of each side of triangle.

Answer 1

[Image Analyst]

help houghlines

 HOUGHLINES Extract line segments based on Hough transform.
    LINES = HOUGHLINES(BW, THETA, RHO, PEAKS) extracts line segments
    in the image BW associated with particular bins in a Hough 
    transform.  THETA and RHO are vectors returned by function HOUGH.
    Matrix PEAKS, which is returned by function HOUGHPEAKS,
    contains the row and column coordinates of the Hough transform 
    bins to use in searching for line segments. HOUGHLINES returns
    LINES structure array whose length equals the number of merged
    line segments found. Each element of the structure array has
    these fields: 
 
       point1  End-point of the line segment; two-element vector
       point2  End-point of the line segment; two-element vector
       theta   Angle (in degrees) of the Hough transform bin
       rho     Rho-axis position of the Hough transform bin
 
    The end-point vectors contain [X, Y] coordinates.
 
    LINES = HOUGHLINES(...,PARAM1,VAL1,PARAM2,VAL2) sets various
    parameters. Parameter names can be abbreviated, and case 
    does not matter. Each string parameter is followed by a value 
    as indicated below:
 
    'FillGap'   Positive real scalar.
                When HOUGHLINES finds two line segments associated
                with the same Hough transform bin that are separated
                by less than 'FillGap' distance, HOUGHLINES merges
                them into a single line segment.
 
                Default: 20
 
    'MinLength' Positive real scalar.
                Merged line segments shorter than 'MinLength'
                are discarded.
 
                Default: 40
 
    Class Support
    -------------
    BW can be logical or numeric and it must be real, 2-D, and nonsparse.
 
    References
    ----------
    Rafael C. Gonzalez, Richard E. Woods, Steven L. Eddins, "Digital
    Image Processing Using MATLAB", Prentice Hall, 2003
 
    Example
    -------
    % Search for line segments corresponding to five peaks in the Hough 
    % transform of the rotated circuit.tif image. Additionally, highlight
    % the longest segment.
 
       I  = imread('circuit.tif');
       rotI = imrotate(I,33,'crop');
       BW = edge(rotI,'canny');
       [H,T,R] = hough(BW);
       imshow(H,[],'XData',T,'YData',R,'InitialMagnification','fit');
       xlabel('\theta'), ylabel('\rho');
       axis on, axis normal, hold on;
       P  = houghpeaks(H,5,'threshold',ceil(0.3*max(H(:))));
       x = T(P(:,2)); 
       y = R(P(:,1));
       plot(x,y,'s','color','white');
 
       % Find lines and plot them
       lines = houghlines(BW,T,R,P,'FillGap',5,'MinLength',7);
       figure, imshow(rotI), hold on
       max_len = 0;
       for k = 1:length(lines)
         xy = [lines(k).point1; lines(k).point2];
         plot(xy(:,1),xy(:,2),'LineWidth',2,'Color','green');
 
         % plot beginnings and ends of lines
         plot(xy(1,1),xy(1,2),'x','LineWidth',2,'Color','yellow');
         plot(xy(2,1),xy(2,2),'x','LineWidth',2,'Color','red');
 
         % determine the endpoints of the longest line segment 
         len = norm(lines(k).point1 - lines(k).point2);
         if ( len > max_len)
           max_len = len;
           xy_long = xy;
         end
       end
 
       % highlight the longest line segment
       plot(xy_long(:,1),xy_long(:,2),'LineWidth',2,'Color','cyan');
 
    See also HOUGH, HOUGHPEAKS.

    Documentation for houghlines
       doc houghlines

[Surabhi A S]

I have tried using hough transform but did not get the output and also was not able to find the edge length

[Image Analyst]

Then use

help imdistline

 imdistline - Distance tool
    An imdistline object encapsulates a Distance tool, which consists of an
    interactive line over an image, paired with a text label that displays
    the distance between the line endpoints.

    Creation
      h = imdistline
      h = imdistline(hparent)
      h = imdistline(___,x,y)

    Input Arguments
      hparent - Handle to parent object
        handle
      x - x-coordinates of endpoints
        2-element numeric vector
      y - y-coordinates of endpoints
        2-element numeric vector

    Properties
      Deletable - ROI can be deleted
        true (default) | false

    Object Functions
      getLabelVisible        - Return visibility of text label of Distance tool
      setLabelVisible        - Set visibility of text label of Distance tool
      getLabelHandle         - Return handle to text label of Distance tool
      getLabelTextFormatter  - Return format of text label of Distance tool
      setLabelTextFormatter  - Set format used to display text label of Distance tool
      getAngleFromHorizontal - Return angle between Distance tool and horizontal axis
      getDistance            - Return distance between endpoints of Distance tool

    See also Line, drawline

    Introduced in Image Processing Toolbox before R2006a
    Documentation for imdistline
       doc imdistline

[Surabhi A S]

Can you please explain on the coding part as well? for the edge length detection and highlighting part

[Image Analyst]

If you don't understand the coding in the examples, just try running them. I think you'll understand what they're saying after you've run the examples.

[Surabhi A S]

Example is fine but I am not able to do it according to my needs.. so i asked you explaination on coding part.

As the edges are not straight and they are curvy.. These functions are not able to detect the complete edge

[Image Analyst]

then maybe you'd find drawline better. I don't have any demos for that but I do for drawpoint, drawpolygon, drawcircle, and drawrectangle. It will be straightforward and easy for you to make the slight changes to one of them to use drawline instead.

[Surabhi A S]

I did not find the function in MATLAB 2018

[Image Analyst]

Can you upgrade? drawline started in R2018b.

[Surabhi A S]

Oh ok. I was able to create a code but not using any of them. Can you pls help out? I am not getting the triangle numbers in the image. I'm not sure on what is wrong in the code.

clc;
close all;
clear all;
 
% Perform edge detection
edgeImage = imread('smooth_imgprocessed_in_matlabcode.jpg');
 
% Convert the edged image to logical
binaryImage = logical(edgeImage);
 
% Perform blob analysis to detect individual triangles
blobAnalysis = vision.BlobAnalysis('AreaOutputPort', false, 'BoundingBoxOutputPort', true, 'CentroidOutputPort', false, 'MinimumBlobArea', 100);
bbox = step(blobAnalysis, binaryImage);
 
% Calculate edge lengths for each triangle
numTriangles = size(bbox, 1);
edgeLengths = zeros(numTriangles, 3);
 
% Create a copy of the image for displaying triangle numbers
outputImage = edgeImage;
 
for i = 1:numTriangles
    % Extract the bounding box coordinates for the current triangle
    x = bbox(i, 1);
    y = bbox(i, 2);
    width = bbox(i, 3);
    height = bbox(i, 4);
    
    % Extract the triangle region from the edge image
    triangleImage = binaryImage(y:y+height-1, x:x+width-1);
    
    % Perform Hough Transform on the triangle image
    [H, theta, rho] = hough(triangleImage);
    
    % Set the minimum line length and gap parameters for `houghpeaks`
    minLineLength = 30;
    maxLineGap = 5;
    
    % Detect peaks in the Hough Transform
    peaks = houghpeaks(H, 3, 'Threshold', 0.3*max(H(:)), 'NHoodSize', [5 5]);
    
    % Extract lines using Hough Transform
    lines = houghlines(triangleImage, theta, rho, peaks, 'FillGap', maxLineGap, 'MinLength', minLineLength);
    
    % Calculate the edge lengths of the triangle
    if length(lines) == 3
        for j = 1:3
            edgeLengths(i, j) = norm(lines(j).point1 - lines(j).point2);
        end
        
        % Display the triangle number on the image
        textPosition = [double(x), double(y)] + [10, 10];
        outputImage = insertText(outputImage, textPosition, num2str(i), 'TextColor', 'red', 'FontSize', 12);
    end
end
 
% Display the edge lengths of each triangle
for i = 1:numTriangles
    disp(['Triangle ', num2str(i), ' Edge Lengths: ', num2str(edgeLengths(i, :))]);
end
 
% Display the image with triangle numbers
imshow(outputImage);
title('Image with Triangle Numbers');

[Image Analyst]

% Demo to show how drawline can be used to draw lines and measure distances on the image.
clc;    % Clear the command window.
close all;  % Close all figures (except those of imtool.)
clear;  % Erase all existing variables. Or clearvars if you want.
workspace;  % Make sure the workspace panel is showing.
format long g;
format compact;
fontSize = 16;

%===============================================================================
% Display demo image in a maximized window.
rgbImage = imread('peppers.png');
hFig1 = figure;
imshow(rgbImage);
hFig1.WindowState = 'maximized';
hFig1.NumberTitle = "off";
hFig1.Name = 'Demo by Image Analyst';

%===============================================================================
% Ask user to draw line over the image.
uiwait(helpdlg('Draw a line.  Click mouse button down, drag, then release mouse button.'));
button = 'Yes';
lineCount = 0;
xy = zeros(1, 4);
while contains(button, 'Yes', 'IgnoreCase',true)
    % User draws a line.  It exits as soon as they lift the mouse.
    hLine = drawline();
    % Get the coordinates in the form [xLeft, yTop, width, height].
    roiPosition = hLine.Position; % Column 1 is the x values, column 2 is y.
    lineCount = lineCount + 1;
    % Save this coordinate.  Format is [x1, x2, y1, y2]
    xy(lineCount, :) = roiPosition(:)';
    % Delete the ROI object.
    delete(hLine);
    % and replace it with a colored line in the graphical overlay.
    hold on;
    plot(xy(lineCount, 1:2), xy(lineCount, 3:4), '.-', 'LineWidth', 2, 'MarkerSize', 30);
    caption = sprintf('You have drawn %d lines so far.', lineCount);
    title(caption, 'FontSize', fontSize);
    % Ask user if they want to click another.
    message = sprintf('%d lines drawn so far.\nDraw another line?', lineCount);
    button = questdlg(message, message, 'Yes', 'No -- Quit', 'Yes');
    if contains(button, 'Quit','IgnoreCase',true)
        break;
    end
end
hold off;
% Get the lengths of the lines.
deltaX = xy(:, 2) - xy(:, 1);
deltaY = xy(:, 4) - xy(:, 3);
lineLengths = sqrt(deltaX .^ 2 + deltaY .^ 2);
meanLength = mean(lineLengths); % Get mean length of all the lines.
caption = sprintf('You drew %d lines.\nMean Line Length = %.1f pixels.', lineCount, meanLength);
fprintf('%s\n', caption)
title(caption, 'FontSize', fontSize);
% Show coordinates in the command window.
xy

%===============================================================================
% Show line lengths as a bar chart.
hFig2 = figure;
hFig2.NumberTitle = "off";
hFig2.Name = 'Demo by Image Analyst';
subplot(1, 2, 1);
bar(lineLengths);
yline(meanLength, 'Color', 'r', 'LineWidth', 2);
grid on;
xlabel('Line Number', 'FontSize', fontSize);
ylabel('Line Length in Pixels', 'FontSize', fontSize);
caption = sprintf('Lengths of the %d Lines You Drew', lineCount);
title(caption, 'FontSize', fontSize);

% Show histogram of the line lengths.
subplot(1, 2, 2);
histogram(lineLengths);	% Take histogram and automatically plot bar chart of line length distribution.
xline(meanLength, 'Color', 'r', 'LineWidth', 2);
grid on;
xlabel('Line Length in Pixels', 'FontSize', fontSize);
ylabel('Count', 'FontSize', fontSize);
caption = sprintf('Mean Line Length = %.1f pixels.', meanLength);
title(caption, 'FontSize', fontSize);
hFig2.WindowState = 'maximized';

[Surabhi A S]

Thanks for the demo. Is there any way in which the length would be automatically displayed, without the user drawing the line. As i ahev amny samples I can't keep on drawing the line to find the length. So I would like the code itself detect the length automatically, I think I have got half of the code done which I have mentioned in earlier comment.