Morphology and regionprops for Counting and Measuring Components on a Robotic Assembly Line

Project info

• Field of studyRobotics and Mechatronics
• LevelMaster's thesis (S2)
• Method typeImage Processing
• Estimated build time± 5 working days
• Toolsstrel, imopen, imclose, regionprops (Image Processing Toolbox)
• PriceUS$28 - US$172

Research goal

Apply morphological operations to the component image while tuning the structuring element size and the minimum object area to assess processing quality.

Variables

Example result

• 4Objects found
• 4Objects true
• 423Mean area (px)
• 0.644Mean eccentricity
• 0.882Mean solidity
• 80.67Pixel error removed (%)

Morphological opening and closing cut the pixel error of Component image from 6.5% to 1.2% and recover 4 objects (true count 4) with a mean area of 423 px and solidity of 0.882.

Computed from illustrative synthetic data, not from a real study.

About Morphology and regionprops

Morphological operations (erosion, dilation, opening, closing) clean and reshape binary images, and regionprops measures properties of each object such as area and perimeter. They are used in particle, cell, and quality inspection analysis.

Steps

1. 1Prepare a binary image from segmentation
2. 2Choose a structuring element with strel
3. 3Apply opening to remove small noise
4. 4Apply closing to fill small holes
5. 5Label objects and measure with regionprops
6. 6Show the property table and plot object boundaries

Assumptions to check

• Structuring element size matches object and noise scale
• The binary image from segmentation is reasonably good
• Pixel-to-real-unit calibration is known if physical sizes are wanted

Reading and reporting the results

Report the object count, area statistics, and other properties with unit calibration. Explain the structuring element choice and its effect on object size.

Example MATLAB code

A short starting script with this project's variable names. The delivered version is adapted to your data and fully commented.

rng(3);
[xx, yy] = meshgrid(1:200);
bw0 = ((xx-50).^2 + (yy-50).^2 < 20^2) | ((xx-130).^2 + (yy-60).^2 < 25^2) | ...
      ((xx-100).^2 + (yy-140).^2 < 30^2);
component_image = bw0;
component_image(rand(200) < 0.02) = true;                     % salt noise
component_image(rand(200) < 0.02) = false;                    % pepper holes

se = strel('disk', 3);
opened = imopen(component_image, se);
cleaned = imclose(opened, se);
cleaned = bwareaopen(cleaned, 50);
structuring_element_size = bwlabel(cleaned);

props = regionprops(structuring_element_size, 'Area', 'Perimeter', 'Eccentricity', 'Centroid');
T = struct2table(props);
T.Circularity = 4*pi*T.Area ./ T.Perimeter.^2;
disp(T(:, {'Area', 'Perimeter', 'Eccentricity', 'Circularity'}));
fprintf('Objects: %d, mean area: %.1f px\n', height(T), mean(T.Area));

figure;
subplot(1,2,1); imshow(component_image); title('Noisy binary');
subplot(1,2,2); imshow(cleaned); hold on;
for k = 1:height(T)
    plot(T.Centroid(k,1), T.Centroid(k,2), 'r+', 'MarkerSize', 10);
end
title('After opening and closing');

Packages and prices

Choose how much help you need. Each higher package includes everything in the lower ones. Prices exclude 11% VAT.

• Payment: 50% down payment before work starts, balance after the deliverables are handed over
• Tax: Prices exclude 11% VAT
• Data and academic integrity: Results are computed from your own research data; you write and stand behind your manuscript, we help with the analysis and your understanding of it
• Revisions: Up to 2 revisions are included in the price (Package 3: 3 rounds)

Frequently asked questions