Unlocking Genetics: Building an Iris Color Detection System with Raspberry Pi

Tweet

Published August 17, 2026

Computer vision has revolutionized how we interact with the physical world, enabling machines to see, analyze, and interpret visual data in real time. While facial recognition and object detection are commonplace, a fascinating intersection of biology, genetics, and embedded systems has emerged: using a Raspberry Pi to detect iris color and explore the genetic mechanisms of eye color inheritance.

By combining high-resolution digital imaging with robust computer vision libraries, makers and educators can construct a sophisticated tool that determines iris phenotype and links it directly to genomic data. Here is an in-depth look at how such a system works, the hardware required, and the underlying technology that brings this DIY science project to life.

The Biological and Technical Link: Why Iris Color Matters

Historically, eye color was taught as a simple Mendelian trait (brown is dominant, blue is recessive). However, modern genetics reveals that eye color is polygenic, determined by multiple genes, primarily OCA2 and HERC2, which control the production and distribution of melanin within the iris stroma.

An iris color detection system bridges the gap between digital imaging and genetic prediction models (such as the IrisPlex system used in forensics). By capturing high-definition images of the eye, processing the pixels to isolate the iris, and analyzing color distribution, a Raspberry Pi can categorize phenotypes and assist in demonstrating how genetic traits manifest physically.

The Hardware Stack

To capture the intricate details of the human iris, high-quality optical components are crucial. Standard low-cost webcams often lack the resolution and focal control required for macro-imaging. The ideal hardware stack includes:

• Single-Board Computer: A Raspberry Pi 4 (4GB or 8GB) or the newer Raspberry Pi 5. These boards offer the computational headroom required for real-time image processing and machine learning inference.
• Camera Module: The Raspberry Pi Camera Module 3 or the High-Quality (HQ) Camera. The Camera Module 3 is particularly useful due to its powered autofocus capabilities, allowing the system to lock onto the eye at close range.
• Lens selection (for HQ Camera): A 6mm CS-mount lens or a specialized macro lens to capture high-density pixel detail of the iris.
• Illumination: A diffuse, low-intensity LED ring light. Proper lighting is critical to illuminate the iris evenly, minimize shadows, and avoid harsh pupil constriction or uncomfortable glare for the subject.

The Computer Vision Pipeline

The core of this project lies in the software pipeline, typically written in Python and utilizing powerful libraries like OpenCV and MediaPipe. The processing pipeline can be broken down into four distinct phases: image acquisition, eye localization, iris segmentation, and color analysis.

1. Eye Localization

Before analyzing color, the software must locate the eye within the camera's field of view. Using MediaPipe Face Mesh, a lightweight framework designed by Google, the system can instantly identify key facial landmarks. By tracking the specific coordinates of the eyelids and pupil, the software crops the frame to focus solely on the ocular region, discarding irrelevant background data.

2. Iris Segmentation

Once the eye region is isolated, the next challenge is separating the iris from the pupil (the dark center) and the sclera (the white of the eye). This is achieved using geometric segmentation techniques:

• Grayscale Conversion and Thresholding: The cropped image is converted to grayscale to find the dark pupil, which serves as the center point.
• Hough Circle Transform: OpenCV's HoughCircles function is applied to detect the circular boundaries of both the inner pupil and the outer iris ring.
• Masking: An annular (donut-shaped) mask is created. This mask isolates the iris tissue while excluding the black pupil and any reflections or white sclera surrounding the eye.

3. Color Space Conversion and Extraction

Analyzing color in the standard RGB (Red, Green, Blue) color space is notoriously unreliable due to its sensitivity to lighting changes. To resolve this, developers convert the isolated iris pixels into the HSV (Hue, Saturation, Value) or CIELAB color space.

In HSV, the "Hue" channel represents the pure color, independent of brightness ("Value") and intensity ("Saturation"). By generating a histogram of hue values from the masked iris region, the software can classify the dominant color profile. For instance, blue eyes will exhibit a high density of pixels within a specific hue range, while brown and green eyes will shift further along the spectrum.

# Sample Python snippet for color space conversion
import cv2
import numpy as np

# Load the cropped iris region
iris_crop = cv2.imread('iris.jpg')

# Convert to HSV color space
hsv_iris = cv2.cvtColor(iris_crop, cv2.COLOR_BGR2HSV)

# Calculate histogram for Hue channel
hist_hue = cv2.calcHist([hsv_iris], [0], None, [180], [0, 180])

Interpreting Genetic Inheritance

With the dominant iris color classified, the Raspberry Pi can run an educational genetic simulation. Users can input the eye colors of their parents, and the software—using Mendelian or more complex polygenic algorithms—can calculate and display the statistical probability of their eye color inheritance.

This interactive component makes the project incredibly valuable for STEM education, allowing students to see concrete visual evidence of genetic expression combined with coding and modern hardware development.

Overcoming Common DIY Challenges

When building this system, makers often run into a few common hurdles:

• Corneal Reflection (Glint): Light sources reflect off the curved surface of the eye, creating bright white spots that skew color data. Utilizing polarized filters over the camera lens and LED light source can dramatically reduce this glare.
• Eyelid Occlusion: Eyelids and eyelashes often cover the top and bottom of the iris. The software must be calibrated to analyze only the lateral (left and right) segments of the iris ring, which are less likely to be obscured.
• Varying Ambient Light: To maintain consistency, the system should be housed in a small, hooded enclosure that shields the eye from ambient room lighting, ensuring the LED ring light is the only active light source.

Conclusion

The Raspberry Pi iris color detection project is a stellar showcase of how low-cost, accessible single-board computers can be used for sophisticated scientific exploration. By combining Python, OpenCV, and careful optical engineering, developers can build a tool that makes complex biological concepts tangible. Whether used as a classroom demonstration tool or a stepping stone toward advanced biometric systems, this project highlights the endless possibilities when hardware and software intersect.