支持人工智能解决方案解决停车问题

Author:Linda Duffy

随着有限数量的停车位和越来越多的汽车，与停车相关的挫败感和不便正在增长。认为这是一个问题，Manuela Rasthofer，首席执行官TerraLoupe GmbH，启动了一个项目combine artificial intelligence and orthorectified aerial imageryto create an accurate inventory of available parking lots and spaces throughout Germany.

Mapping available parking

Parking a car can be a stressful and time-consuming activity, and in the future, autonomously navigated vehicles will be searching for spaces without a driver to help. The need for high-definition digital maps that accurately measure and identify all types of objects, including parking spaces, is quickly becoming a reality.

TerraLoupe GmbH is a technology start-up based in Munich, Germany, that focuses on combining geodata and computer analytics in innovative ways. Starting with high-resolution orthoimagery, TerraLoupe appliesmachine learning algorithms to detect and measure objects in the physical world, such as buildings, roads, and trees, to create data-rich 3D models.

“为了解决越来越多的停车问题，我们想看看使用空中图像和人工智能检测和评估停车场是否可行，”says Rasthofer.“通过自动化功能和数字内容的提取，我们认为我们可以大大减少创建地图所​​需的时间，而不会牺牲准确性。”

Acost-effective method of creating digital mapsis particularly interesting to Tier one automotive suppliers and original equipment manufacturers (OEMs) to support the autonomous navigation industry; however, many other industries can make use of the information as well.

HxGN Content Program delivers

基于Martinsburg的15厘米GSD数据，由检测到的车道标记创建的语义车道模型

2014年，HxGN Content Programbegan collecting speculative off-the-shelf矫形图像美国，欧洲部分地区以及加拿大人口稠密的地区创建一个可供客户使用的数据库。目标是获取无云的30厘米分辨率，4频段图像over less populated areas, and15-cm resolutionover metro areas with apopulation greater than 50,000。

Through the HxGN Content Program, TerraLoupe obtained15-cm resolution orthoimagery of Berlin测试其内部开发的对象识别算法。柏林的最初工作eight weeks to train the algorithmsto accurately identify and categorise parking spaces, followed by just三天分析和生成地图for all of Germany.

“通过HXGN内容程序访问图像allows us todownload the geographic locationswe need, and then在新数据上训练我们的算法,”explained Rasthofer.“每个国家 /地区独有的建筑，基础设施和道路系统总是有轻微的差异。我们检查每个对象的置信区间，并重新检查低百分比。当我们纠正错误时，算法继续学习和改进，直到达到非常高的准确性水平为止。”

Theaerial orthoimages available through the HxGN Content Programgo through a rigorous QA/QC process to ensure delivery of survey-grade images.“ HXGN内容计划最适合我们客户的需求areas of autonomous driving, parking assistance, and loss reportsfor insurance/reinsurance companies,”says Rasthofer.“We also successfully deliver情报有关的基础设施、公用事业、railways以及其他目的。”

Machine learning expedites accurate mapping

TerraLoupe’s project shows thathigh-resolution aerial orthoimages combined with machine learning可以有效地用于提取数字内容。停车分析提供了有用的信息，例如停车场的位置，入口和出口以及可以适合每个批次的不同类别的汽车数量（紧凑，中型，大型）。城市规划师，送货人员，出租车司机和拥挤零售区的顾客都可以从这种改善的停车情报中受益。

“可用性高分辨率，高准确图像确定我们在哪里开始该项目；但是，我们打算在整个欧洲进行此分析，因为数据通过Hexagon获得，我们希望将服务扩展到美国，”bob体育报道says Rasthofer.“Overall our goal is to efficiently extract all types of objects andcreate a complete digital environment。”

Obtainingaerial imagery is fasterand moreefficientthan terrestrial methods, allowing more frequent updates, which is crucial for many applications. Hexagon’s global operations generate widespread availability of imagery and good business partnerships with data providers to continue to meet thegrowing demand for digital maps。