Parking Seeker

• Alex Dai
• Alfredo Ruiz
• Dara Chea

In today’s urban environment vehicles have become used as a staple in transportation. One problem that arises when driving a vehicle is parking, specifically the act of locating parking spaces. The current services for providing parking information are limited to only parking lots without mentioning the availability of spaces along the streets. The objective of this project is to develop a system that collects data, specifically length in between parked vehicles, along a particular street using imaging sensors and reports back the availability based on the parameters defined by the user. The data will be used to determine the number of available spaces on the street for parking. The system will also take into account the input of the type of vehicle and use that information to determine the appropriate number of spaces available according to the length of the vehicle. In order to fulfill this task, we have created a device and named it the Parking Seeker. It has been developed to accomplish a more efficient and safer way to park. Using a Raspberry Pi and a camera we are able to capture frames and detect cars using OpenCV and haar cascade. We trained the haar cascade with over a thousand photos of cars in order to create a vector to recognize the cars in the frame capture by Raspberry Pi camera module. To demonstrate the functionality, we have created a controlled scale replica of the environment of and urban street side with ideal lighting and atmospheric conditions. Furthermore, to properly show the recognition function and distancing function of our device we have used model cars in proportion to the test environment. The models and environment that are presented were the same ones that were used to test our device during development.