Background 3d bldg. models and features
AAM were commissioned to create 3D building models across 500 square kilometres in metropolitan Perth. Due to project timelines and budget, AAM opted for an Artificial Intelligence (AI) based solution. This approach saved the client time and money, especially when compared to conventional mapping methods.
AAM Approach
Due to the project costs and time constraints, AAM decided to adopt an Artificial Intelligence (AI) based approach for the 3D Mapping to create a spatial ‘Digital Twin’ of the project area. AAM’s experience with using AI and Machine Learning (ML) algorithms suggested that we could quickly adapt our Machine Learning engine for this project. AAM were confident that previously developed Machine Learning Algorithms and workflows for transforming high resolution aerial imagery into 3D vector data could be quickly adapted and ‘trained’ for this project.
Challenges
The targeted area of work included a mix of area densities covering residential, commercial, and industrial locations. Rural areas and vineyard developments were also included. On initial investigation, it was unclear how effective an AI-based approach would be in extracting features from this wide variety of landscapes.
Solution
AAM utilised 15 cm stereo aerial imagery covering 500 square kilometres across the Perth metropolitan region. Machine Learning algorithms were applied to automatically extract 2D vector layers from these datasets. It was completed with the same level of accuracy as could be achieved manually by a GIS professional.
To support the 3D requirement, AAM generated a Digital Surface Model (DSM) from the stereo aerial imagery. This DSM enabled AAM staff to transform the 2D data into 3D data, including precise building heights.
The following features were extracted:
- Multilevel Buildings
- Roads
- Railways
- Bridges
- Water bodies
- Swimming pools
- Trees
Results
It took just 12 days to extract, check and deliver the required features. Without the use of AI, it would have taken up to 10,500 hours to achieve the same results, or just over 6 months using a 10-person mapping team.
AAM estimates that the AI based process provided a project saving of 42% over conventional processes.
Based on the success of this project, we know that an AI-based approach for feature extraction can be adopted by organisations looking to increase outputs whilst minimising costs. Traditional mapping methods are very linear: the larger the area, the higher the cost. In contrast, AI pushes the unit cost down as the size of the area goes up.” (Michael De Lacy, Mapping Lead, AAM Group)
Once the project outputs were finalised, several stakeholder groups reviewed the feature extractions. Their feedback was positive and comprised the following:
- Good consistency and accuracy in the modelling
- Smooth footprints and better definition of shape
- Data appears “ready to use”
- Building footprints broken down into bulk components, indicating a complex shape and varying heights instead of a single outline with a single spot height
- Bridge polygons were presented well
- Successful categorisation of water types
- Very comprehensive and good symbology
The use of AI in projects like this demonstrates clear advantages. Spatial data can be maintained easily and frequently. AAM can help businesses and organisations scale their spatial operations to suit their needs in an agile way”. (Michael De Lacy, Mapping Lead, AAM Group)