Kit Introduction

The Myduino AIoT Computer Vision Kit is a complete, hands-on development platform designed to help learners and developers build real AI-powered monitoring systems. This kit integrates OpenCV for image processing, YOLO for real-time object detection, and MQTT for IoT communication, combining AI intelligence with hardware control into one structured, step-by-step learning experience — using real working systems, not simulations.

Throughout this journey, users will transform a camera and a microcontroller into an AI-powered system that can see, process images, detect objects using YOLO, and make intelligent decisions. With OpenCV handling image analysis and MQTT enabling wireless data communication, the system can automatically control physical devices in real time.

From camera input and image processing to AI detection, wireless communication, and hardware automation, learners will experience the full AIoT workflow in a practical, application-driven way — building a complete end-to-end AI system from vision to action.

What is OpenCV

OpenCV (Open Source Computer Vision Library) is a free and open-source library used for image and video processing. It helps computers understand and analyse visual data from cameras, photos, and videos. With OpenCV, you can detect faces, recognise objects, track movement, read QR codes, and apply image effects like blur or edge detection.

It is commonly used with programming languages like Python and C++, and it is popular in AI, robotics, and smart camera projects. In simple words, OpenCV helps computers “see” and understand the world through images and videos.

What is YOLO

YOLO stands for You Only Look Once. It is a real-time object detection system used in computer vision. YOLO can detect multiple objects in an image or video simultaneously, such as people, cars, animals, or other items.

It is fast and accurate because it looks at the image only once and predicts all object locations and classes in a single step. This makes it well-suited for real-time applications such as CCTV systems, self-driving cars, robotics, and AI cameras. YOLO is commonly used together with libraries like OpenCV and deep learning frameworks to build smart vision systems.

What is MQTT

MQTT stands for Message Queuing Telemetry Transport. It is a lightweight communication protocol used for sending data between devices over the internet.

MQTT is commonly used in IoT (Internet of Things) systems because it is simple, fast, and uses very little bandwidth. It works using a publish and subscribe system. Devices (called clients) send messages to a broker, and other devices can subscribe to receive those messages. A popular MQTT broker example is Mosquitto.

MQTT is widely used with devices like ESP32, Raspberry Pi, and other IoT hardware to send sensor data, control devices remotely, and build smart home or industrial monitoring systems.

Why Choose This Kit?

• Complete AI + IoT integration in one kit
• Step-by-step structured exercises included
• Real hardware control (not simulation only)
• Suitable for teaching, training, and prototyping
• Expandable for advanced AIoT projects

Suitable For

• Lecturers and educational institutions
• Final Year Project (FYP) students
• Makers and developers
• Beginners entering the AIoT industry

1-Day Class Exercises

This kit includes 6 progressive practical exercises:

Exercise 1: Image Processing Using OpenCV

1. How to load and validate an image using OpenCV.
2. Converting a BGR image into Grayscale for easier processing.
3. Applying Canny Edge Detection with adjustable thresholds.
4. Understanding how image-processing parameters affect the final output.

Exercise 2: Object Detection Using YOLO

1. How to initialise and stream video from a USB webcam using OpenCV.
2. Improving performance with reduced input resolution (imgsz).
3. Using cv2.imshow() and keyboard controls to safely exit the program.

Exercise 3: Object Filter via Serial Communication

1. How to extract specific objects (like “person” or “cell phone”) from a list of detections.
2. Turning AI bounding box coordinates into simple strings of text.
3. Using the pyserial library to transmit detection alerts to external devices

Exercise 4: IoT MQTT Integration

1. How to send small, fast data packets over the internet to a “Broker”
2. Connecting your Python script to an ESP32 microcontroller without any physical wires.

Exercise 5: AIoT Traffic Control

1. Writing if/elif statements to handle different detected labels.
2. How to switch between Red, Yellow, and Green LEDs based on real-time vision.
3. Configuring a real ESP32 to subscribe to “Traffic” topics and drive hardware pins.

Exercise 6: AI Braking System

1. How to maintain a “Run” state by default.
2. Using the if/else logic to toggle between MOVE and STOP commands.
3. Programming the system to automatically recover once the trigger (the person) is removed.

Complete AIoT Workflow

Camera ➜ OpenCV Processing ➜ YOLO Detection ➜ Object Filtering ➜ MQTT Communication ➜ ESP32 ➜ Hardware Output

What’s Included in the Kit

Buy From

• Myduino AIoT Computer Vision Kit from Myduino.com