Hough Transform for Detecting Floor and Wall Edges as Indoor Robot Navigation Guides

Project info

• Field of studyRobotics and Mechatronics
• LevelMaster's thesis (S2)
• Method typeImage Processing
• Estimated build time± 5 working days
• Toolshough, houghpeaks, houghlines (Image Processing Toolbox)
• PriceUS$25 - US$159

Research goal

Apply the Hough transform to the corridor image while tuning the angular resolution and the accumulator threshold to assess processing quality.

Variables

Example result

• 3Lines expected
• 3Lines detected
• 0.246Mean angle error (deg)
• 0.381Max angle error (deg)
• 169Peak votes
• 3Segments found

The Hough transform finds 3 of 3 lines in the Corridor image edge image with a mean angle error of 0.25 degrees and a peak accumulator of 169 votes; 3 line segments were extracted.

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

About Hough Transform

The Hough transform detects geometric shapes such as lines and circles by voting in parameter space. It is used to detect road lines, building edges, and machine parts in images.

Steps

1. 1Convert the image into a binary edge image
2. 2Compute the transform with hough
3. 3Find accumulator peaks with houghpeaks
4. 4Extract line segments with houghlines (FillGap and MinLength)
5. 5Display the angle and distance of each line
6. 6Draw detected lines over the image

Assumptions to check

• Edges are reasonably clear and not too noisy
• The number of peaks sought is estimated sensibly
• FillGap and MinLength match the line scale

Reading and reporting the results

Report the accumulator parameters, the number of lines detected, and each line's angle and distance versus true values if known. Explain sensitivity to the peak threshold.

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(5);
BW = false(200);
BW(60, 20:180) = true;                              % horizontal line
for k = 20:180, BW(k, k) = true; end                % 45-degree diagonal
BW(20:180, 150) = true;                             % vertical line
corridor_image = BW | (rand(200) < 0.003);                   % add speckle noise
angular_resolution = 3;                                         % number of peaks sought

[H, T, R] = hough(corridor_image);
P = houghpeaks(H, angular_resolution, 'Threshold', ceil(0.3*max(H(:))));
lines = houghlines(corridor_image, T, R, P, 'FillGap', 5, 'MinLength', 40);
fprintf('Lines found: %d\n', numel(lines));
for k = 1:numel(lines)
    fprintf('Line %d: theta = %d deg, rho = %d, length = %.1f px\n', k, lines(k).theta, ...
        lines(k).rho, norm(lines(k).point1 - lines(k).point2));
end

figure;
subplot(1,2,1);
imshow(imadjust(rescale(H)), [], 'XData', T, 'YData', R, 'InitialMagnification', 'fit');
axis on; axis normal; hold on; plot(T(P(:,2)), R(P(:,1)), 'rs');
xlabel('theta'); ylabel('rho');
subplot(1,2,2); imshow(corridor_image); hold on;
for k = 1:numel(lines)
    xy = [lines(k).point1; lines(k).point2];
    plot(xy(:,1), xy(:,2), 'g', 'LineWidth', 2);
end

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