Imgage Processing Tools

Background

During high school, I was the first in my robotics team to successfully implement a real-time vision system for object detection and tracking. This system became a critical part of our competition robot, allowing precise detection of circular and colored targets.

My work was recognized both by my school and the local municipality, where I received an honor for my contributions to technological innovation in youth robotics. This project taught me practical computer vision, problem-solving under pressure, and how persistence leads to real results.

---

Overview

This repository contains several Python scripts for computer vision tasks in robotics:

• Real-time color detection using HSV filtering
• Contour analysis and visualization
• Angle and distance estimation from camera to target
• Modular design for testing and competition purposes
• GUI to visualize results and adjust parameters in real-time

Among all scripts, GenericVision.py is the main module: a robust, fully-featured vision system that can detect multiple targets, compute angles and display annotated outputs in real-time.

---

Files

File	Description
GenericVision.py	Main vision system. Supports multiple targets, angle estimation, and real-time display.
circular_objects_recognition.py	Detects circular objects in video streams, with contour visualization.
contour_simple_recognition.py	Simplified contour-based detection for testing or small targets.
recognition+distance_for_comp.py	Competition-ready script: detects targets, computes distance, displays annotations.
consts.py	Stores constants like FOV, target size, HSV ranges, and camera parameters.
hsvimage.png	An image used for Explaining HSV in dynamic trackbar window
README.md	This file.

---

Installation

Install dependencies:

pip install opencv-python numpy imutils

---

Usage

Run the main vision module:

python GenericVision.py

Controls:

• Space: Pause / resume
• Esc or q: Quit

Other scripts can be used for testing or simpler vision tasks:

python circular_objects_recognition.py
python contour_simple_recognition.py
python recognition+distance_for_comp.py

Adjust HSV values and camera exposure in consts.py for your specific environment.

---

Notes

• Designed for high school robotics competitions
• Works best under consistent lighting conditions
• Scripts are modular: you can combine functions to create a custom vision pipeline

---

Crazy Eyes: Face and Eye Overlay

This repository also includes a fun real-time face and eye manipulation script, designed as an experimental exercise in computer vision:

• Detects faces and eyes using Haar cascades (haarcascade_frontalface_default.xml and haarcascade_eye.xml)
• Switches eyes between detected faces in real-time
• Optional framing around faces and eyes to visualize detection
• Provides a live GUI with instructions displayed at the bottom of the window

Controls while running:

• Space: Show / hide switched eyes
• Esc: Exit the program
• 1: Toggle framing around faces and eyes

It complements the robotics vision scripts by showing how object detection and manipulation can be applied to different types of targets (faces/eyes) beyond competition objects.