This document explains how to use the Parking Lot Classification — USA pretrained model available on ArcGIS Living Atlas of the World. The model is used to detect and classify pixels of parking areas from high-resolution aerial RGB imagery.
Parking lots in the United States occupy significant land area, particularly in urban and suburban areas. Using parking spaces for solar panel installation is a growing trend, known as solar parking lots or solar carports. While solar energy has made substantial progress in the United States, there is still untapped potential. By installing solar panels on parking structures, it is possible to use this space for solar energy generation without requiring additional land. By doing so, they not only provide shade for parked vehicles but also generate clean energy and reduce the carbon footprint of buildings and facilities. They can also be combined with electric vehicle (EV) charging infrastructure to estimate the potential demand for electric vehicles, which can be powered by the solar panels installed in the parking lot, promoting the adoption of clean transportation and reducing reliance on fossil fuels and further enhancing sustainability. But traditionally, parking areas are manually digitized and classified, which is a labor- and time-intensive task. Automating the task using deep learning models for parking space classification and solar panel capacity calculation outperforms traditional methods in terms of efficiency, accuracy, scalability, adaptability, real-time monitoring, and integration with renewable energy goals.
The use of GeoAI for parking space classification and solar panel installation capacity calculation can have potential applications in urban planning, land use optimization, renewable energy deployment, and sustainable transportation and contribute to the country's renewable energy goals.
License requirements
To complete this workflow, the following are the license requirements:
- ArcGIS Desktop—ArcGIS Image Analyst extension for ArcGIS Pro
- ArcGIS Enterprise—ArcGIS Image Server with raster analytics configured.
- ArcGIS Online—ArcGIS Pro or Professional Plus user type.
Model overview
This model has the following characteristics:
- Input—8-bit, 3-band high resolution (30 centimeters -1.2 meters) imagery. For detecting small sized parking lots, higher resolution imagery is highly recommended.
- Output—-Feature layer representing classified parking spots.
- Compute—This workflow is compute-intensive, and a GPU with minimum CUDA compute capability of 6.0 is recommended.
- Applicable geographies—This model is expected to work well in the United States.
- Architecture—-This model uses the MMSegmentation based DeepLabV3Plus model architecture implemented in ArcGIS API for Python.
- Accuracy metrics—This model has an average precision score of 0.75 and recall of 0.68.
- Limitations
- The model is expected to work well on commercial paved parking lots.
- The model might get confused with paved surface having similar reflectance.
Access and download the model
Download the Parking Lot Classification — USA pretrained model from ArcGIS Living Atlas of the World. Alternatively, access the model directly from ArcGIS Pro, or consume it in ArcGIS Image for ArcGIS Online.
- Browse to ArcGIS Living Atlas of the World.
- Sign in with your ArcGIS Online credentials.
- Search for Parking Lot Classification — USA and open the item page from the search results.
- Click the Download button to download the model.
You can use the downloaded .dlpk file directly in ArcGIS Pro or upload and use it in ArcGIS Enterprise. Additionally, you can fine-tune the pretrained model if necessary.
Release notes
The following are the release notes:
| Date | Description |
|---|---|
| July 2023 | First release of Parking Lot Classification — USA |