Smart Camera Parking System With Auto Parking Spot Detection
Tuan T. Nguyen, Mina Sartipi
TL;DR
The paper tackles scalable smart parking by addressing automatic parking spot localization and occupancy detection without extensive manual labeling. It introduces PakLoc, an automatic localization method using a deformable DETR-based vehicle detector, plus PakSke for rotating and resizing bounding boxes to align with true parking-spot orientations, and PakSta for automatic occupancy identification by reusing PakLoc detections. On the PKLot dataset, PakLoc achieves an AR75 of up to 94.25% with PakSke, while PakSta reaches AP75 of 93.6% without requiring manual labeling of parking spots, demonstrating strong performance against baselines. Collectively, these components enable a more cost-efficient, scalable Smart Parking System by reducing labeling effort and providing real-time occupancy information across diverse camera setups.
Abstract
Given the rising urban population and the consequential rise in traffic congestion, the implementation of smart parking systems has emerged as a critical matter of concern. Smart parking solutions use cameras, sensors, and algorithms like computer vision to find available parking spaces. This method improves parking place recognition, reduces traffic and pollution, and optimizes travel time. In recent years, computer vision-based approaches have been widely used. However, most existing studies rely on manually labeled parking spots, which has implications for the cost and practicality of implementation. To solve this problem, we propose a novel approach PakLoc, which automatically localize parking spots. Furthermore, we present the PakSke module, which automatically adjust the rotation and the size of detected bounding box. The efficacy of our proposed methodology on the PKLot dataset results in a significant reduction in human labor of 94.25\%. Another fundamental aspect of a smart parking system is its capacity to accurately determine and indicate the state of parking spots within a parking lot. The conventional approach involves employing classification techniques to forecast the condition of parking spots based on the bounding boxes derived from manually labeled grids. In this study, we provide a novel approach called PakSta for identifying the state of parking spots automatically. Our method utilizes object detector from PakLoc to simultaneously determine the occupancy status of all parking lots within a video frame. Our proposed method PakSta exhibits a competitive performance on the PKLot dataset when compared to other classification methods.