What it shows
Proxy attendance is a problem in classes and workshops. A fingerprint module identifies a person on the spot and a log records the time. The project builds a stand-alone attendance box and a Python script that converts the log into a monthly report, with attention to consent and data protection.
The science
The module has its own processor. It captures an image, extracts minutiae points (ridge endings and bifurcations) and stores a template in its flash against an ID number. A match returns the ID and a confidence score. The microcontroller never sees the fingerprint image. The RTC supplies the time and the log line records ID, date and time on the SD card.
Materials
Cost: under Rs 5,000. Many items come from home or from waste; spare parts such as LEDs, motors and wire are in the store.
- An Arduino Mega or ESP32
- An R307 or AS608 fingerprint module
- A DS3231 real-time clock module with its coin cell
- A microSD card module
- A 16x2 I2C LCD and a buzzer
- A 5 V power adapter that is a sealed ready-made USB type
- A box and a keypad or buttons for enrol and delete
Safety
For this project: Use only sealed USB power. Collect fingerprints only from volunteers with written consent and explain how templates are stored. Never store images, only the module templates, and delete them at the end of the project.
Never use mains electricity: power models with cells, batteries or a low-voltage adapter. Sharp and hot tools, flames and lab chemicals need an adult or your teacher beside you. Keep magnets, button cells and small parts away from young children, and wash your hands after making.
All the safety rulesStep by step
11 steps. Do them in order, and test each part before you go on to the next. Write what you notice in a notebook as you go.
Connect the sensor over a serial port, the RTC over I2C and the SD module over SPI.
Install the Adafruit fingerprint library and run the enrol example for one volunteer.
Write a menu: hold the enrol button to add a new ID, and otherwise the device waits for a finger.
Match the finger, read the ID and confidence and show the name from a table of IDs stored in a CSV file on the SD card.
Log date, time and ID to attendance.csv and prevent double logging within 5 minutes.
Add a buzzer tone and an LCD message for accepted and rejected fingers.
Enrol 10 volunteers with 2 fingers each and test 20 attempts per person at different angles and with dry or damp fingers.
Calculate the false rejection rate and try to find the false acceptance rate by having volunteers try each other’s IDs.
Write a Python script that reads the CSV and produces a monthly summary with first-in and last-out times.
Add an admin function to delete templates and test it.
Write the report with the architecture, results, security and privacy risks, and a consent form.
What you should see
The box recognises enrolled volunteers in about one second, logs correct times, and the Python summary matches the number of test attendances. False rejections occur mostly with dry or dirty fingers and false acceptance should be near zero.
Viva questions
A teacher or a judge will ask how it works. Try to answer each question before you open it, then compare.
1What are minutiae?
Points in a fingerprint where ridges end or split, used for matching.
2What is FRR and FAR?
False rejection rate is genuine users refused, and false acceptance rate is wrong people accepted.
3Why does the module store a template, not an image?
A template is smaller and cannot easily be turned back into the fingerprint image.
4What privacy steps are needed?
Written consent, local storage only, no sharing and a way to delete data.
Take it further
Send the log to a school server over Wi-Fi in a daily batch and show parents a monthly attendance percentage without fingerprint data.
Change one thing, measure what happens, and add a table or chart of your readings: that is what turns a model into your own investigation.