The research funded by Grosvenor Engineering Group, Enviro Building Services and the NSW Department of Industry, achieved its three main goals and worked well with SMART's Digital Living Lab's project to provide an Internet of Things (IoT) Network to create smarter living in buildings.

"The outcomes were two-fold and are changing how the heating, ventilation and cooling of buildings can be used more efficiently in the built environment," senior research fellow and team leader Dr Rohan Wickramasuriya said.

"An accurate real time people counting solution for indoor environments that respects privacy is now available for building managers to utilise.

"A deep learning-based indoor temperature forecasting algorithm has been developed which provides a great alternative to the traditional approach that requires detailed information about a building's construction, fit-out and modeled occupancy.

"Training this algorithm for a new building is straightforward, hence it will cut time and costs when a forecasting model is required to predict indoor temperatures."

Grosvenor national sustainability manager Rod Kington welcomed the results.

"Making buildings more efficient, environmentally sustainable and improving the building value is at the core of what Grosvenor does," he said.

"Applying the UOW's scientific research and real-world industry experience to the built environment brought solutions to challenges. This partnership has enabled us to leverage the knowledge across our extensive building network."

Products from the research that are now being used by the industry include a cost-effective accurate people counter, combining off the shelf components with powerful machine learning software.

Dr Wickramasuriya added that the research undertaken into deep neural networks was a future driver of artificial intelligence in smart buildings.

"We started off with a deep neural network architecture called YOLO (You Only Look Once) that can detect objects in real time scenes," he said.

Read more: Hundreds turn out to Wollongong's National Book Relay

"Using transfer learning, we custom trained the YOLO algorithm to detect people in the images, reviewed annotated images and then evaluated the accuracy of the algorithm. The test set accuracy of the algorithm was 92 per cent which is an excellent result, given the existing solution's accuracy is often only 65 per cent.

"In addition, to forecast room temperature, we used another deep network architecture called Long Short Term Memory (LSTM). LSTM models were compared against classical time series forecasting models and vastly outperformed the classical models from an accuracy perspective."