Author:
Gyulai-Nagy Zoltán-Valentin
Affiliation:
Department of Computer Science, Babeş-Bolyai University, 1, M. Kogălniceanu Street, 400084, Cluj-Napoca, Romania
Keyword(s):
Clustering, SOM Clustering, HDBSCAN, K-Means, Real-Time Application, Automation, Parking Lot Space Detection, Feature Extraction.
Abstract:
The paper presents an approach for detecting symmetrical clusters in noisy data, using parking space detections as a real-world example. The paper proposes a plug-and-play solution that uses camera systems to automatically detect parking spaces and provide metrics about availability and accuracy. The approach uses clustering algorithms and image detection for data acquisition and mapping, which can be easily adapted to any application that requires geometrical data extraction. The paper also presents the different phases involved in mapping parking spaces and the challenges that need to be addressed. Overall, the proposed approach can benefit both parking lot administrators and drivers by providing real-time information on available parking spaces and reducing emissions, fuel costs, traffic, and time spent searching for a spot.