NewsSmart HVAC Project Uses Computer Vision to Optimize Cooling in Each Zone...

Smart HVAC Project Uses Computer Vision to Optimize Cooling in Each Zone Individually

Category articles

Jallson Suryo has developed an innovative smart HVAC project utilizing computer vision technology in order to reduce energy waste and optimize cooling efficiency. Traditional HVAC systems often fail at adapting to changing occupancy or activity levels, leading to energy losses; but Suryo’s project aims to revolutionize how individual zones are cooled by providing additional cooling based on specific metrics tracked through computer vision technology.


Suryo created a 1:50 scale model of an office environment to demonstrate his concept, featuring four rooms and human figurines. He captured 79 images with Edge Impulse and used annotated bounding boxes around each person to train a customized object detection model using FOMO (Faster R-CNN Object Detection with MobileNet) algorithm – this allowed precise tracking and analysis of individuals within the model.

Suryo took his project one step further by installing OpenMV firmware onto an Arduino Nicla Vision board and using real-time detections of individuals within his model. This firmware served as the underlying software controlling camera module functions; using OpenMV in tandem with this board enabled Suryo to harness computer vision’s potential instantly and view detections instantly.

Suryo completed his development process by crafting an Arduino library to incorporate the trained model. She seamlessly integrated this library into a sketch that communicated with an Arduino Nano peripheral board via I2C protocol, relaying information on people per quadrant to allow dynamic adjustment of one of four 5V DC fans allowing efficient temperature regulation within each zone while all pertinent information was conveniently displayed on an OLED screen for monitoring and control purposes.

Jallson Suryo’s proof of concept marks a revolutionary leap forward for HVAC technology. Integrating computer vision with intelligent control systems, his smart HVAC project offers promise to both reduce energy waste and increase cooling efficiency in buildings. For those eager to dive deeper, Suryo has provided an extensive writeup on his Edge Impulse documentation page so both enthusiasts and professionals alike can fully realize this groundbreaking proof of concept.

What Is an HVAC System

HVAC stands for Heating, Ventilation and Air Conditioning and refers to equipment and technologies used to control temperature, humidity, air quality and ventilation within buildings or spaces. An HVAC system refers to equipment used for managing these parameters to achieve thermal comfort while maintaining suitable indoor air quality for occupants – these components could include furnaces, heat pumps, air conditioners, ventilation systems ductwork thermostats – among others – these HVAC systems can be found both residential and industrial buildings alike to create comfortable indoor environments.

What Is OpenMV Firmware

OpenMV firmware refers to the software or program running on an OpenMV Cam board, a low-cost microcontroller-based camera module designed for computer vision applications. The OpenMV Cam comes equipped with both a microcontroller and camera sensor; its firmware regulates their functionality and operation.

OpenMV firmware offers a suite of preprogrammed functions and algorithms designed for image processing and computer vision tasks, enabling users to interact with camera modules, capture images or video, perform various image analysis operations such as color detection, motion detection, face recognition etc. and interact with their camera modules directly.

Firmware typically offers a user-friendly programming environment that enables users to configure camera settings, write scripts or programs to control camera behavior, and process images or video in real-time.

OpenMV firmware is open-source, meaning its source code can be freely modified and tailored by users in the community. This enables developers and enthusiasts to extend the functionality of an OpenMV Cam and build custom computer vision applications using this firmware as their starting point.

Electronics and Telecommunications engineer with Electro-energetics Master degree graduation. Lightning designer experienced engineer. Currently working in IT industry.