Sep, 2021 - By WMR
Roofpedia utilizes a fully convolutional neural network (deep learning) to allow researchers and practitioners to examine how urban areas throughout the world are greening their rooftops and installing solar installations.
As cities throughout the world become more crowded, the need to improve and expand existing areas grows. Cities are focusing more on how underutilized rooftop spaces could help with climate change, agricultural production, and other issues. Sustainable roofs, such as those with plants and solar panels, can help cities reduce their carbon footprint, but although studies have been conducted to evaluate their possibilities, few have followed cities' overall performance.
Roofpedia, an automated tool that uses satellite images to track how rooftops around the world adopt solar panels and/or plants, was developed by Abraham Noah Wu, NUS Master of Architecture graduate, and Dr. Filip Biljecki, Professor at the National University of Singapore School of Design and Environment.
Roofpedia is an open roof database that contains data from 1 million rooftops in 17 morphologically and geographically diverse cities. Owing to its high rankings in both areas covered and a number of green roofs, Zurich received a score of 100. With an index score of 86, Las Vegas came out on top for solar roof adoption. This could be owing to the geographical area's high solar potential.
Each city has its distinct qualities, and the precise value of rooftop flora or solar panels is highly dependent on the city's structure and architecture. The city's geolocation and macroclimate also play a role. Green roofs are more difficult to maintain in drier climates, while solar roofing may not be cost-effective in rainy and dark climates.
Roofpedia's data on 1 million rooftops has been made publicly available, and the research team hopes that other researchers will join them in expanding the database by recording other cities or integrating other environmental factors.