Model for Detecting Parking Spaces in Urban Streets

Today, parking space detection models in urban streets are recognized as tools for optimizing the use of public spaces, improving traffic, and increasing citizen satisfaction. In this study,
a model has been designed that estimates the parking space status using variables such as street congestion and parking violations. Data were collected from different urban areas, including traffic plan areas, air pollution control areas, and other areas, at different times of the day and night to analyze traffic patterns and user behavior. Information on parking spaces and violations was collected in the field and data on traffic congestion was collected through Google Map software. The relationship between different hours, traffic congestion and traffic violations was analyzed through linear regression, the relationship between parking space and hours of the day was examined through the Friedman test, and the difference in parking space on holidays and non-holidays was determined with the Wilcoxon test. Also, the relationship between parking space and type of use was compared through the independent chi-square test and the Phi agreement coefficient. The results showed that the proposed model is able to estimate parking space with high accuracy. These results can be very useful for citizens, mobile navigation software and parking management systems.