Rolling-Shutter Velocity Measurement Project

ISEF Category: Physics and Astronomy

Subcategory: Mechanics  ·  Difficulty: Advanced  ·  Setup: School Lab  ·  Time: Full Year

The Hook

Your phone camera can lie about motion in a useful way. Fast moving objects bend, tilt, or shear because the sensor reads each row at a slightly different time. That glitch can become a speed sensor. If you can decode it well, you can measure rotation with surprising precision.

What Is It?

A rolling shutter camera does not capture the whole frame at once. It scans the image row by row. If an object moves while that scan happens, the object looks slanted or warped. That shape records motion the way a clock records time, but only if you know how to read it.

Think of it like a photocopier moving across a page while the page itself slides sideways. The copied lines will not line up straight. In this project, you use that distortion as data. You connect the amount of tilt or spacing change in the image to the speed of a fan blade, a spinning propeller, or a moving fluid feature such as a vortex edge.

You also need a reference model. That means you need to know your phone’s readout timing, camera frame rate, and the geometry of the moving object. Once you build that bridge, you can turn weird-looking images into real velocity estimates.

Why This Is a Good Topic

This makes a strong science fair topic because the core idea is measurable, visual, and tied to a real engineering problem. You are not just taking pictures. You are turning camera artifacts into a motion sensor and testing how accurate that sensor can be. The project connects to smart imaging, robotics, and motion analysis, and you can learn calibration, error analysis, and model testing without needing a full lab setup.

Research Questions

• How does phone model affect rolling-shutter-based speed estimates for the same rotating object?
• What is the effect of object brightness and contrast on the accuracy of rolling-shutter velocity measurements?
• Does using a longer exposure or different frame rate change the error in measured rotational speed?
• To what extent does blade shape or propeller geometry change the relationship between image distortion and true speed?
• Which image feature, edge tilt, spacing shift, or stripe angle, gives the most precise velocity estimate?
• How does camera angle relative to the motion axis affect the measured speed error?
• What is the effect of using a calibration object with known rotation speed on the final percent error?

Basic Materials

• Smartphone with rolling-shutter video mode.
• Tripod or phone clamp.
• Rotating object, such as a desk fan or small motor with visible blades.
• Printed high-contrast target or marker tape for the moving object.
• Measuring tape or ruler.
• Notebook or spreadsheet for recording results.
• Light source with steady brightness.
• Basic safety goggles.

Advanced Materials

• High-speed reference camera or optical tachometer.
• Function generator and motor controller.
• Stroboscope or LED pulse source for comparison measurements.
• Calibration rig with known angular speed.
• Laser tachometer for validation.
• Computer with image analysis software.
• Balanced test rotors or propeller assembly.
• Water channel or clear tank for vortex imaging.

Software & Tools

• ImageJ: Measures edge angles, spacing changes, and frame-to-frame distortion in video stills.
• Tracker: Tracks motion points and helps compare image-based speed estimates with reference motion.
• Python: Lets you extract frames, fit calibration curves, and calculate percent error across trials.
• OpenCV: Automates edge detection and image measurements from rolling-shutter video frames.
• Google Sheets: Organizes trials, graphs calibration data, and compares prediction error.

Experiment Steps

1. Define the motion pattern you will measure, and choose one object geometry that gives a clean image signal.
2. Characterize your camera’s timing behavior, including frame rate and any rolling-shutter constraints that affect the model.
3. Build a calibration plan that links image distortion features to a trusted speed reference.
4. Choose the image feature you will measure, then decide how you will extract it consistently from every trial.
5. Plan controls that separate true motion signals from lighting changes, blur, and camera angle effects.
6. Set your analysis method before collecting final data, including how you will report uncertainty and percent error.

Common Pitfalls

• Using auto-exposure or auto-focus changes between clips, which shifts contrast and makes the distortion harder to measure consistently.
• Photographing the rotor from a changing angle, which mixes true speed changes with perspective distortion.
• Choosing a motion target with weak edges, which makes the tilt or stripe pattern too noisy for reliable analysis.
• Comparing rolling-shutter estimates to a shaky reference method, which hides whether the camera model or the validation tool caused the error.
• Ignoring sync between lighting flicker and camera readout, which can create fake patterns that look like motion.

What Makes This Competitive

A stronger version of this project does more than show that distortion exists. It builds a calibrated model, tests several object types, and reports uncertainty cleanly. You can stand out by comparing multiple image features, validating against an independent speed reference, and checking whether your method holds across different phones or lighting setups. Careful error analysis matters as much as the motion physics.

Project Variations

• Test the method on ceiling fan blades, then compare accuracy across blade counts and blade angles.
• Adapt the analysis to drone propellers and study how propeller size changes distortion sensitivity.
• Use water-surface vortices or spinning dye patterns and compare image-based speed estimates with a reference sensor.

Learn More

• MIT OpenCourseWare, High-Speed Imaging related lecture notes: Search MIT OpenCourseWare for courses on imaging, motion analysis, or experimental methods.
• NASA camera calibration and imaging resources: Search NASA for guidance on image sensors, optics, and measurement from imagery.
• NIH PubMed: Search review articles on rolling-shutter imaging, motion estimation, and camera calibration.
• ImageJ documentation: Find official guides for measuring angles, edges, and regions of interest in video frames.
• OpenCV documentation: Use the official tutorials to learn frame extraction, edge detection, and simple motion analysis.