Urban Rat Activity Tracking
ISEF Category: Animal Sciences
Ready to Turn This Idea Into a Real Project?
This guide was put together with the help of AI research tools to give you a solid starting point.But a competitive science fair project lives in the details: refining your research question, fine-tuning your variables, analyzing your data, and presenting your findings like a seasoned scientist.
For next steps tailored to your interests, skill level, and timeline, work one-on-one with a MehtA+ mentor. Learn more about MehtA+ Science & Engineering Research Mentorship →
Subcategory: Other · Difficulty: Intermediate · Setup: Home Setup · Time: 1 to 2 Months
The Hook
Urban rats leave more clues than most people realize. A phone camera can catch those clues, if you know what to count and when to count it. That makes this project a mix of animal behavior, data science, and public-health tracking. You are not just watching rats, you are turning messy street video into evidence.
What Is It?
This project studies how to measure rat activity from short sidewalk-camera clips. Think of it like building a scoreboard for a game that happens fast, at night, and in a cluttered stadium. The goal is to decide which video cues, like crossings, pause points, or group size, best turn raw footage into a reliable activity measure.
A citizen-science protocol means regular people can help collect or label the data. Your smartphone app prototype can guide them, standardize the inputs, and reduce guesswork. In plain terms, you are asking, "How can a phone help people count rats the same way every time?"
Why This Is a Good Topic
This is a strong science fair topic because you can measure it, compare methods, and improve the protocol step by step. It connects to real urban wildlife monitoring, sanitation, and public-health planning. You can learn video annotation, app design, sampling, and basic statistics without needing a university lab.
Research Questions
- How does video length affect the number of rat detections per clip? ?
- What is the effect of time of night on rat activity counts from sidewalk-camera footage? ?
- Does adding a simple counting guide improve agreement between student rat annotations? ?
- To what extent do camera angle and sidewalk width change detection accuracy? ?
- Which activity metric, first appearance, total crossings, or dwell time, best tracks repeated rat visits? ?
- How does clip quality, such as blur or low light, affect the false positive rate? ?
Basic Materials
- Smartphone or tablet with video playback
- Laptop or desktop computer
- Spreadsheet software
- Free video annotation tool
- Sample sidewalk-camera clips
- Headphones
- Notebook for protocol notes
- Digital timer
- Internet access for citizen-science research
Advanced Materials
- High-frame-rate smartphone or action camera
- Dedicated field laptop
- External hard drive for clip storage
- ImageJ or Fiji for frame-by-frame review
- Python with OpenCV and pandas
- Jupyter Notebook
- Statistical software such as R or JASP
- Optional GPS logger for site metadata
- Access to institutional review guidance if human contributors label data
Software & Tools
- ImageJ: Reviews clips frame by frame and helps you mark rat appearances consistently.
- Python: Automates clip sorting, annotation checks, and summary counts.
- OpenCV: Supports motion-based detection and basic video processing.
- Google Sheets: Tracks labels, compares rat counts, and computes agreement scores.
- JASP: Runs simple statistics without paid software.
Experiment Steps
- Define the one rat activity metric you will measure first, so your project stays focused.
- Choose a small set of clip features to test, such as light level, camera angle, or crowding.
- Build a labeling guide that tells volunteers exactly what counts as a rat event.
- Plan a comparison between human labels and app-assisted labels to check consistency.
- Set up a standard way to summarize counts across clips and sites.
- Decide how you will test whether the protocol works better than a no-guide baseline.
Common Pitfalls
- Counting the same rat twice when it leaves and reenters the frame, which inflates activity scores.
- Mixing clips from very different neighborhoods without tracking site context, which makes the comparison unfair.
- Using blurry or overexposed footage as if it were clear footage, which raises false detections.
- Letting volunteers interpret rat behavior differently, which drops agreement between labels.
- Testing too many app features at once, which makes it hard to tell what improved the result.
What Makes This Competitive
A stronger version of this project would not just count rats, it would prove that one counting method is better than another. You could compare human labels, app-assisted labels, and simple computer vision on the same clips. If you add inter-rater agreement, error analysis, and site-level controls, the project starts to look like real field-methods research. That kind of design shows you understand both the animal behavior and the measurement problem.
Project Variations
- Use park-edge camera clips instead of sidewalk footage to compare rat activity near green space.
- Compare nighttime clips with early morning clips to test whether activity patterns shift by hour.
- Test whether a motion-based app or a manual checklist gives more consistent rat counts from the same videos.
Learn More
- USGS National Wildlife Health Center: Search for urban wildlife monitoring methods and rodent ecology resources.
- NIH PubMed: Search review articles on urban rats, rodent behavior, and zoonotic risk.
- NOAA Citizen Science resources: Look for plain-language guides on building reliable volunteer protocols.
- MIT OpenCourseWare, Introduction to Computer Vision: Use it to learn the basics of frame analysis and detection.
- OpenCV documentation: Read the free tutorials for video loading, tracking, and motion detection.
- Google Scholar: Search recent papers on urban rat activity, citizen science, and video-based monitoring.