Vision Lab - Computer Vision Application

VisionLab is a real-time computer vision application built with Qt (QML + C++) and OpenCV. The project focuses on a camera pipeline, live image rendering, and an extensible architecture for adding vision modules (such as motion detection, object detection, face detection) in a clean and modular way.

The current implementation prioritizes performance, stability, and clean architecture, with detectors designed to be developed and integrated as separate modules or projects.

• Features

• Live camera capture using OpenCV

• Real-time frame rendering in QML via a custom QQuickImageProvider

• Clean separation between UI, camera management, and vision logic

• Motion detection module (lightweight and CPU-friendly)

• Dark/idle frame handling when no camera stream is available

• Extensible design for adding future detectors (object, text, etc.)

• Technologies Used

• C++ (C++17)
• Qt 6
• Qt Quick (QML)
• Qt Core / GUI
• OpenCV
• CMake
• QML Image Provider
• Multithreading (std::thread)

• Models Used in This Project

• Face Detection

Model:ResNet SSD Face Detector (Caffe) Files: deploy.prototxt, res10_300x300_ssd_iter_140000.caffemodel Framework: OpenCV DNN Purpose: Detect human faces in the live camera stream.

• Object Detection

Model: YOLOv4-Tiny Files: yolov4-tiny.cfg, yolov4-tiny.weights, coco.names Framework: OpenCV DNN (Darknet backend)

• Motion Detection

Model: Background Subtraction (No ML model) Algorithm: OpenCV Background Subtractor (e.g. MOG2) Framework: OpenCV Purpose: Detect motion regions in the camera feed.

• Setup & Development

• Prerequisites

Make sure you have the following installed:

Qt 6.x Qt Quick Qt Widgets

OpenCV (built with DNN disabled is fine for this project)

CMake ≥ 3.16 C++17 compatible compiler MSVC / GCC / Clang

• OpenCV Setup

Ensure OpenCV is correctly installed and accessible.

Example (Windows):

# Clone the repository
git clone https://github.com/Muhammedsuwaneh/vision-lab.git

# Navigate to the project folder
cd vision-lab

set(OpenCV_DIR "C:/opencv/build")
find_package(OpenCV REQUIRED)

Verify:

opencv_world or core modules are found

Camera access works (cv::VideoCapture)

mkdir build
cd build
cmake ..
cmake --build .

Or open directly in Qt Creator and configure with CMake

• Run the Application

After building:

Run from Qt Creator or Execute the generated binary from the build folder Make sure:

A webcam is connected Camera permissions are granted (Linux/macOS)

• Version

1.0.0

• Licensed

Under MIT - Copyright 2025/2026