Rapid Urban Database Generator
2022-26-1220
05/26/2022
- Event
- Content
- In recent times, there is an increasing demand for Computer Vision and 3D simulation applications in the field of aviation. Instances of which include virtual pilot training, mission planning, guidance and tracking, etc. The current 3D database generation process demands enormous manual effort and cost to create elements with precise detail. Seamless integration of new content and detailed information are the key features in 3D database generation, which often demands continuous and recurring work. Our idea is to develop a solution based on Machine Learning and computer vision technologies for the process of rapid database generation - which shall reduce the manual effort requirements and subsequently the cost. Using a series of high resolution images, we train a machine learning model to automatically detect and classify various objects of interest (roads, buildings) on unseen image data. This automation process can eventually speed up the process of large database generation and can enable the user to visualize and access categorically classified data in a short duration. Additionally, as the processing time is reduced, new content integration and maintenance of an updated database can be done rapidly.
- Citation
- Sawant, R., "Rapid Urban Database Generator," SAE Technical Paper 2022-26-1220, 2022, .