Parking space state detection system of visual parking space

Abstract

According to the technical scheme, the parking space state detection system comprises a processing module, a camera shooting module, a vehicle real-time module, a detection module and a parking space judgment module; the processing module comprises a global model uni;, the camera shooting module shoots a parking space and displays parking space coordinates in the global model unit; the parking space real-time module is arranged on a vehicle, according to the vehicle real-time module, a coordinate origin is generated with the center of a rear axle of the vehicle, and the vehicle mileage of the vehicle in a global coordinate system is monitored; a probe in the detection module detects a parking space and generates probe coordinates in the global coordinate system; the parking space judgment module judges the parking space to determine whether the parking space is an empty parking space or not is judged; and then the parking space is displayed in the vehicle real-time module, so that a driver can obtain accurate parking space information. The parking space state detection system of the visual parking space has the effects of accurately detecting the parking space and assisting the vehicle in parking.

Description

technical field [0001] The invention relates to the technical field of parking space detection, in particular to a visual parking space state detection system. Background technique [0002] In recent years, with the rapid development of the economy, the number of cars has increased rapidly, but the limited number of parking spaces has made it difficult to find parking spaces. The development of parking space detection technology has solved the problem of finding parking spaces to a certain extent. problem. [0003] At present, there are many methods of detecting parking spaces, among which the method of detecting through fisheye cameras is the most common way at present. This detection method is generally detected through traditional image algorithms or deep learning algorithms. Deep learning can detect whether parking spaces are It is an empty parking space, but the detection effect of deep learning mainly depends on the learning samples, that is, the model samples of the ...

AI Technical Summary

Problems solved by technology

[0003] At present, there are many methods of detecting parking spaces, among which the method of detecting through fisheye cameras is the most common way at present. This detection method is generally detected through traditional image algorithms or deep learning algorithms. Deep learning can detect whether parking spaces are It is an empty parking space, but the detection effect of deep learning mainly depends on the learning samples, that is, the model samples of the parking lot. However, the parking space scene is extremely complex and changeable, and it is difficult to support the empty parking space detection of all parking spaces through deep learning methods; traditional images The algorithm is to identify the parking space by detecting the angle between the parking space lines of the parking space. Generally, the state of the parking space cannot be detected, and it cannot accurately judge whether the parking space is empty. Moreover, the fisheye camera has a limited field of view and can only be used to identify 3 parking spaces within the parking space area. The range of about one meter cannot meet more common parking space scenarios. Therefore, the parking space detection method based on the fisheye camera has low accuracy in judging the parking space, and cannot well meet the functional requirements of assisted parking.

Images