A Student Smart Parking Solution using Raspberry Pi
Răzvan Vîlceanu1, Andrei Ternauciuc2, Mihai Onița3
1Răzvan Vîlceanu*, Communication, Politehnica University Timisoara, Timisoara, Romania.
2Andrei Ternauciuc, Communication, Politehnica University Timisoara, Timisoara, Romania.
3Mihai Onița, Communication, Politehnica University Timisoara, Timisoara, Romania.
Manuscript received on October 12, 2019. | Revised Manuscript received on 22 October, 2019. | Manuscript published on November 10, 2019. | PP: 2727-2733 | Volume-9 Issue-1, November 2019. | Retrieval Number: A4912119119/2019©BEIESP | DOI: 10.35940/ijitee.A3912.119119
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The concept of a smart city was born out of the need to provide improved quality of life to citizens in various fields, such as environment, governance, education, economy, infrastructure, transportation, traffic, and parking. There are many countries, many players and many projects already finished or underway. The city of Timisoara has already taken the first steps towards a smart city. It has high potential in this regard, but for the moment is still lacking in some key areas. For example, Timisoara has excessive traffic and usually overcrowded parking lots without some efficient monitoring solutions. As a result of this fact, a team made up of a Master student in his second year of studies, and two faculty coordinating members propose a parking management solution based on a very restrictive budget. The low-cost prototype we developed for testing purposes contains a mockup parking platform with 8 parking spaces, one Raspberry Pi Zero W micro-computer, and a camera module attached to the Raspberry Pi. The idea is to broadcast live feed towards a computing platform tasked with identifying the available parking spots and consequently transmit this information in real-time, via the Internet, to interested parties. We tested and compared two detection algorithms based on corner detection of the parking spots, and the detection of circles placed inside the parking spot, respectively. These algorithms were implemented using open-source technologies such as Python version 3.6.5, OpenCV 3.4.0 specialized modules and functions, digital processing (NumPy), chart plotting (Matplotlib) and data formatting (JSON). The main functions we used were: cv2.imread, cv2.imshow, cv2.imwrite, cv2.VideoCapture, cv2.line, cv2.circle, cv.puttext, cv2.rectangle, cv2.elipse, cv2.blur, cv2.gaussianblur, cv2.canny, cv2.houghlinesp, cv2.houghcircles, cv2.hough_gradient and cv2.selectROI. The solution we proposed is focused on improving the way the Politehnica University Timisoara’s staff and students approach parking. It can reduce traffic congestion and the level of CO2 or other pollutants around University buildings and surrounding areas. Overall traffic would be greatly diminished because the driving lanes would mainly be used for transportation, instead of low-speed cruising in search of a free parking space.
Keywords: Smart Parking, Smart Cities, Computer Vision, Raspberry Pi, Python, Open CV
Scope of the Article: Smart Cities