Histogram Equalisation for Improving Robot Camera Images under Dim Lighting Conditions

Project info

• Field of studyRobotics and Mechatronics
• LevelUndergraduate thesis (S1)
• Method typeImage Processing
• Estimated build time± 3 working days
• Toolshisteq, adapthisteq, imhist, entropy (Image Processing Toolbox)
• PriceUS$22 - US$134

Research goal

Apply histogram equalisation to the robot camera image while tuning the grey level count and the illumination level to assess processing quality.

Variables

Example result

• 0.046Contrast std before
• 0.289Contrast std after
• 4.29Entropy before (bits)
• 4.223Entropy after (bits)
• 0.236Range before
• 1Range after

Histogram equalisation raises the contrast of Robot camera image from a standard deviation of 0.046 to 0.289 and the intensity range from 0.24 to 1.00; entropy moves from 4.29 to 4.22 bits.

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

About Histogram Equalisation

Histogram equalisation (histeq) spreads image intensities to increase contrast, especially in dark or washed-out images. It is used as preprocessing for medical, satellite, and industrial images before segmentation.

Steps

1. 1Read the image and show its histogram
2. 2Apply histeq for global equalisation
3. 3Apply adapthisteq (CLAHE) for local equalisation
4. 4Compare images and histograms before and after
5. 5Compute contrast measures (standard deviation, entropy)
6. 6Choose the method that does not overly amplify noise

Assumptions to check

• The image has a narrow intensity range so equalisation helps
• Global equalisation can amplify noise in homogeneous areas
• Intensities are no longer valid for photometric measurement after equalisation

Reading and reporting the results

Show images and histograms before and after plus quantitative contrast values. Explain that contrast improvement is visual and can alter the original intensity values.

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(4);
[xx, yy] = meshgrid(1:256);
scene = 0.5*double(((xx-90).^2 + (yy-120).^2) < 50^2) + 0.3*double(xx > 160) + 0.2*yy/256;
robot_camera_image = 0.25 + 0.15*mat2gray(scene) + 0.01*randn(256);          % low-contrast image
robot_camera_image = min(max(robot_camera_image, 0), 1);

he = histeq(robot_camera_image);
ad = adapthisteq(robot_camera_image, 'ClipLimit', 0.02);
grey_level_count = {robot_camera_image, he, ad};
names = {'Original', 'histeq', 'adapthisteq'};
for k = 1:3
    I = grey_level_count{k};
    fprintf('%-12s std = %.3f, entropy = %.3f\n', names{k}, std(I(:)), entropy(I));
end

figure;
for k = 1:3
    subplot(2,3,k); imshow(grey_level_count{k}); title(names{k});
    subplot(2,3,k+3); imhist(grey_level_count{k});
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