Autonomous 3D Modeling of House Roofs from Remote Sensing Data

根据遥感数据对房屋屋顶进行自主 3D 建模

• 批准号: 531017-2018
• 负责人: Wong, Alexander
• 金额: $ 1.82万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=668774
• 关键词: Autonomous; 3D; Modeling; House; Roofs

Aerialytic aims to solve an industry-wide lead-qualifying challenge in the solar energy industry, through which we**will enable both homeowners and solar sales companies to overcome barriers associated with utilizing solar**panels, thereby expanding solar energy adoption within Canada and globally.**Currently, solar sales companies are spending 60% of their lead acquisition budget in the solar sales ecosystem.**A solar sales agent needs to manually qualify every single homeowner that inquires about solar panels; this**includes creating a custom design for every home that they believe would reap savings from using solar energy.**This lengthy process is required even for the large number of homeowners (40%) that do not qualify due to**factors such as roof shade or size. These sales funnel inefficiencies result in high operational costs for solar sales**companies and a poor customer experience for homeowners, who must wait up to 2 days to obtain information,**leading to an over 30% drop off-rate due to loss of interest.**Our solution is to automate the qualification process by using tools from Artificial Intelligence. This project will**design and build an end-to-end multi-stage system to automatically extract an accurate roof geometry for a**house from satellite imagery, and provide homeowners with their custom solar design on the spot, along with**information on financial and environmental returns.**Last year over 350,000 homeowners transitioned to solar energy in the US alone; with the help of our technology,**we believe we can amplify this number in the US and across the globe.

Aerialytic旨在解决太阳能行业的全行业领先资格挑战，通过该挑战，我们将使房主和太阳能销售公司能够克服与利用太阳能电池板相关的障碍，从而在加拿大和全球范围内扩大太阳能的采用。目前，太阳能销售公司在太阳能销售生态系统中花费了60%的铅收购预算。太阳能销售代理需要手动资格查询太阳能电池板的每一个房主;这 ** 包括为他们认为可以从使用太阳能中节省费用的每一个家庭创建定制设计。这一漫长的过程是必要的，即使是大量的房主（40%）不符合资格，由于 ** 因素，如屋顶阴影或大小。这些销售漏斗效率低下导致太阳能销售公司的运营成本很高 **，房主的客户体验很差，他们必须等待长达2天才能获得信息，** 由于利息损失，导致超过30%的折扣率下降。我们的解决方案是通过使用人工智能工具来自动化资格认证过程。该项目将 ** 设计和构建一个端到端的多阶段系统，以自动从卫星图像中提取 ** 房屋的准确屋顶几何形状，并为房主提供现场定制的太阳能设计，并沿着 ** 有关财务和环境回报的信息。**去年，仅在美国就有超过350，000名房主过渡到太阳能;在我们技术的帮助下，** 我们相信我们可以在美国和整个地球仪扩大这个数字。