Smartphone Microscope Vibration Isolation Project

ISEF Category: Engineering Technology: Statics and Dynamics

Subcategory: Control Theory  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

Tiny shakes can wreck a microscope image. A table vibration you barely feel can blur fine detail enough to hide cells, fibers, or crystal edges. If you can keep a phone-based microscope steady, you can make the same camera suddenly look much sharper. That makes this a strong control systems project with a real payoff.

What Is It?

This project studies active vibration isolation, which means a system senses motion and pushes back to cancel it. Think of it like noise-canceling headphones, but for shaking instead of sound. Your stage uses voice coils, which are electromagnets that move a mount, and flexures, which are bendable structures that guide motion without loose joints.

The smartphone microscope is your test platform. A phone camera records the image, and you measure sharpness with Laplacian variance, a number that rises when edges stay crisp and falls when the picture blurs. Your control system, running on an STM32 microcontroller, tries to predict and reduce incoming vibration before it ruins the image.

The interesting part is the loop between mechanics, sensing, and software. You are not just building a stand. You are asking how well a feedback or feed-forward controller can improve image quality under real vibration.

Why This Is a Good Topic

This is a good science fair topic because you can measure performance with numbers, not just photos. You can test how control settings, mount stiffness, sensor choice, or vibration frequency change image sharpness. The project connects to camera stabilization, precision instruments, and robotics, so the real-world value is easy to explain. A student can learn control basics, data analysis, and experimental design without needing a full research lab at the start.

Research Questions

• How does active vibration isolation change Laplacian variance in smartphone microscope images?
• What is the effect of different vibration frequencies on sharpness improvement from the controller?
• Does adaptive feed-forward filtering outperform fixed-gain control for image stabilization?
• To what extent does flexure stiffness change the isolation stage's response to table vibration?
• Which sensor placement gives the best prediction of image blur before compensation?
• How does the number of stacked flexure mounts affect resonance and sharpness gain?

Basic Materials

• Smartphone with a microscope clip or phone lens attachment.
• Basic microscope slide set or printed fine-pattern target.
• Small vibration source such as a motor, fan, or speaker-based shaker.
• Rigid base plate for the stage.
• Flexible mount material or simple flexure prototype parts.
• Accelerometer or motion sensor module.
• STM32 development board.
• Voice-coil actuator or small linear actuator.
• Breadboard, jumper wires, and connectors.
• Tripod or fixed phone mount.
• Digital kitchen scale for comparing prototype mass.
• Tape measure or ruler.
• Laptop for data logging and analysis.

Advanced Materials

• Optical breadboard or precision base plate.
• Laser displacement sensor or high-resolution accelerometer.
• Voice-coil actuators matched to the stage load.
• Custom-machined flexure mounts.
• STM32 board with motor driver interface.
• Function generator for controlled vibration input.
• DAQ system for synchronized sensor logging.
• High-magnification calibration target or resolution chart.
• Vibration isolation pads for baseline comparison.
• 3D printer or CNC access for mount prototypes.
• Image analysis setup for batch processing microscope frames.

Software & Tools

• ImageJ: Measures Laplacian variance and other sharpness metrics across image sets.
• Python: Organizes data, plots vibration response, and compares controller settings.
• OpenCV: Processes microscope images and extracts blur or edge features.
• GNU Octave: Helps model controller behavior and test signal filters.
• STM32CubeIDE: Programs the microcontroller and manages embedded control code.

Experiment Steps

1. Define the vibration problem you want to suppress and choose one image metric as your main outcome.
2. Map the system parts, including the sensor, actuator, flexure stage, and phone camera, so you know where each signal enters and leaves.
3. Build a baseline model of how the uncorrected stage behaves under vibration before you add active control.
4. Choose the control strategy you will compare first, such as feed-forward, feedback, or a simple adaptive filter.
5. Plan a calibration method that links motion measurements to image blur so you can turn pictures into numbers.
6. Design comparison tests that separate mechanical improvements from software improvements, then repeat them under different vibration conditions.

Common Pitfalls

• Measuring sharpness from photos taken under changing exposure settings, which makes Laplacian variance hard to compare.
• Testing the controller without a true baseline, which hides whether the stage improved anything at all.
• Ignoring resonance in the flexure stack, which can amplify vibration instead of canceling it.
• Using unsynchronized sensor and camera data, which makes blur prediction weak.
• Tuning the controller only for one vibration frequency, which makes the system look good in one case and fail in others.

What Makes This Competitive

A stronger version of this project does more than prove that the stage works. It compares multiple control strategies, reports transfer functions or frequency response, and uses repeated trials with clear statistics. You can stand out by testing several vibration sources, several sample loads, or several flexure designs, then showing which combination gives the best sharpness gain. A polished project also explains failure modes, not just success cases.

Project Variations

• Test the same isolation idea on a handheld phone microscope, then compare blur reduction while the user moves the device slightly.
• Swap the vibration source for a desktop fan or motor platform, then study which frequencies the controller cancels best.
• Replace Laplacian variance with edge contrast or Fourier-based sharpness, then see whether the conclusions change.

Learn More

• MIT OpenCourseWare: Search for control systems and feedback lectures to learn stability, transfer functions, and controller design.
• NIH PubMed: Search review articles on vibration isolation, image stabilization, and microscope imaging to find prior methods.
• NASA NTRS: Search for papers on active vibration control and adaptive filtering used in precision systems.
• NOAA Education Resources: Look for basic wave and signal material that helps with frequency and filtering ideas.
• ImageJ Documentation: Read the official guides for batch image measurement and sharpness analysis.
• STM32 Documentation: Use the official reference manuals and application notes to understand timers, sensors, and motor control.