A Pipeline and NIR-Enhanced Dataset for Parking Lot Segmentation

TL;DR

This paper presents a new dataset and methodology using Near-Infrared (NIR) imagery to enhance the accuracy of parking lot segmentation from satellite images, leveraging deep learning models and post-processing techniques.

Contribution

Introduces two large datasets with RGB and NIR channels for parking lot segmentation and demonstrates improved accuracy using NIR and post-processing with deep learning models.

Findings

NIR channel improves segmentation accuracy.

Post-processing techniques further enhance results.

Best model achieves 84.9% mIoU and 96.3% accuracy.

Abstract

Discussions of minimum parking requirement policies often include maps of parking lots, which are time consuming to construct manually. Open source datasets for such parking lots are scarce, particularly for US cities. This paper introduces the idea of using Near-Infrared (NIR) channels as input and several post-processing techniques to improve the prediction of off-street surface parking lots using satellite imagery. We constructed two datasets with 12,617 image-mask pairs each: one with 3-channel (RGB) and another with 4-channel (RGB + NIR). The datasets were used to train five deep learning models (OneFormer, Mask2Former, SegFormer, DeepLabV3, and FCN) for semantic segmentation, classifying images to differentiate between parking and non-parking pixels. Our results demonstrate that the NIR channel improved accuracy because parking lots are often surrounded by grass, even though the NIR channel needed to be upsampled from a lower resolution. Post-processing including eliminating erroneous holes, simplifying edges, and removing road and building footprints further improved the accuracy. Best model, OneFormer trained on 4-channel input and paired with post-processing techniques achieves a mean Intersection over Union (mIoU) of 84.9 percent and a pixel-wise accuracy of 96.3 percent.