Review of Research on Vision-Based Parking Space Detection Method

Review of Research on Vision-Based Parking Space Detection Method

Yong Ma, Yangguo Liu, Shiyun Shao, Jiale Zhao, Jun Tang
Copyright: © 2022 |Pages: 25
DOI: 10.4018/IJWSR.304061
Article PDF Download
Open access articles are freely available for download

Abstract

Parking space detection is an important part of the automatic parking assistance system. How to use existing sensors to accurately and effectively detect parking spaces is the key problem that has not been solved in the automatic parking system. Advances in Artificial Intelligence and sensing technologies have motivated significant research and development in parking space detection in the automotive field. Firstly, based on extensive investigation of a lot of literature and the latest re-search results, this paper divides parking space detection methods into methods based on traditional visual features and those methods based on deep learning and introduces them separately. Secondly, the advantages and disadvantages of each parking space detection method are analyzed, compared, and summarized. And the benchmark datasets and algorithm evaluation standards commonly used in parking space detection methods are introduced. Finally, the vision-based parking space detection method is summarized, and the future development trend is prospected.
Article Preview
Top

Introduction

Due to the surge in the number of cars in our country, the number of urban parking spaces is far from meeting the requirements of the existing number of cars. The limited parking space makes roads, streets and parking lots congested. On the other hand, more and more novice drivers are on the road, aggravating this situation and increasing the probability of traffic accidents. In the current complex traffic environment, there are higher requirements for the driver's driving skills.

For the driver, it is a difficult thing for the old driver and the novice driver to park the car. During the parking process, it is not only necessary to move the vehicle back and forth repeatedly, so that the car can park in the parking space accurately, and in the process of parking, always pay attention to the obstacles and passers-by around the vehicle to avoid collision, which is a very difficult problem for the driver, which also puts forward a more difficult problem for the driver's driving level and reaction time. high demands. Through the analysis of the causes of traffic accidents and the accumulation of parking experience in life, it is concluded that there are the following difficulties in the parking process:

  • (1)

    The blind spot of the driver's field of vision. In the car, the driver can only observe the obstacles around the vehicle through the rear-view mirror. Due to the limited information observed in the rear-view mirror, there are blind spots of vision behind the left and right sides of the vehicle, and the image displayed through the rear-view mirror will appear to some extent. The degree of distortion leads to the driver's unclear understanding of the outside environment of the car, which increases the difficulty of parking in place.

  • (2)

    The fault tolerance rate of the driver operating the vehicle is low. Reversing is an essential step in parking, but it is often difficult for novice drivers with inexperienced drivers, and it is necessary to control the steering, accelerator and brake pedal of the vehicle while reversing. Mentioned the difficulty of parking in place.

  • (3)

    The limitation of parking space size. In the limited urban space, the space left for parking spaces is getting smaller and smaller. In addition, the parking positions of some adjacent parking spaces are not standardized, which will further compress the limited parking space, making it even more difficult to park in the space.

A large number of traffic accidents caused by parking and consumers' demands for car safety in the process of parking and warehousing, and with the proposal of automatic driving, automatic parking assistance system has attracted extensive attention of researchers (Frank, 2014; Kageyama, 2004), it is imminent to develop an automatic parking assistance system to reduce the difficulty of parking and improve the safety of the car. In the process of parking in place, the system first uses various sensors to sense the environmental information around the vehicle, eliminates blind spots in the field of vision, and provides real-time early warning when encountering dangerous situations, so as to build a safe parking space for the driver; secondly, the system can replace the driver to control the steering wheel for automatic steering operation, improve driving safety; finally, according to the different parking spaces detected, the corresponding parking strategy will be selected to achieve fast and accurate parking. In the process of developing an automatic parking system, a key issue that needs to be solved is how to use the onboard sensors to correctly detect and locate the parking space. The solutions to this problem can generally be divided into two types, free-space-based parking space detection method and vision-based parking space detection method (Wan et al., 2009).

Complete Article List

Search this Journal:
Reset
Volume 21: 1 Issue (2024)
Volume 20: 1 Issue (2023)
Volume 19: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 18: 4 Issues (2021)
Volume 17: 4 Issues (2020)
Volume 16: 4 Issues (2019)
Volume 15: 4 Issues (2018)
Volume 14: 4 Issues (2017)
Volume 13: 4 Issues (2016)
Volume 12: 4 Issues (2015)
Volume 11: 4 Issues (2014)
Volume 10: 4 Issues (2013)
Volume 9: 4 Issues (2012)
Volume 8: 4 Issues (2011)
Volume 7: 4 Issues (2010)
Volume 6: 4 Issues (2009)
Volume 5: 4 Issues (2008)
Volume 4: 4 Issues (2007)
Volume 3: 4 Issues (2006)
Volume 2: 4 Issues (2005)
Volume 1: 4 Issues (2004)
View Complete Journal Contents Listing