Parking space detection method

Abstract

The invention provides a parking space detection method. The method comprises the following steps: constructing a large-scale, annotated looking-around image data set for a parking space perception problem; improving a target detection framework YOLOv2 framework based on deep learning to complete parking space control point detection; pairing each two detected control points to form point pair combination, performing preliminary screening on the distance constraints, designing a custom model to classify point pairs of parking space control points that satisfy the distance constraints on the basis of a shallow model AlexNet based on a classification model of deep convolutional neural network, and judging the type of a parking space and the entrance line direction to complete parking space deduction. The parking space detection method provided by the invention is an important component of an autonomous parking system in unmanned driving, the position of the parking space can be detectedonly by cameras around the vehicle body, and a technical support is provided for the autonomous parking. The parking space detection method provided by the invention is not only high in detection precision, but also is high in response speed, and is stable and reliable.

Description

technical field [0001] The invention belongs to the field of auxiliary parking for computer vision and intelligent driving, and relates to a parking space detection method, especially a parking space detection method based on deep learning technology. Background technique [0002] With modern humans' pursuit of automation and convenient life, all walks of life are developing in the direction of automation and intelligence. As the main means of transportation for people, automobiles have an extremely broad market, and their automation and intelligence are also an important topic in today's academic circles and automotive engineering circles. The automobile industry has proposed the concept of smart cars, and strives to realize fully automated and driverless cars. Among them, the autonomous parking system, as a key technology in the field of smart cars, has gradually become a popular research direction for major automakers. For drivers, especially beginners, the limited park...

AI Technical Summary

Problems solved by technology

The work of Suhr and Jung (Suhr et al., Full-automatic recognition of various parking slot markings using a hierarchical tree structure, 2013) uses the Harris corner detector to detect the corner points, find the intersection points of the parking slot lines based on these corner points and To determine the parking space, the limitation of this method is that it relies on the accuracy of Harris corner detection, which is not stable enough

Images