Parking space detection method, device and equipment and storage medium

Abstract

The embodiment of the invention discloses a parking space detection method, device and equipment and a storage medium, and the related embodiments can be applied to various scenes such as cloud technology, artificial intelligence and smart traffic, and are used for improving the accuracy of parking space acquisition and improving the parking efficiency. The method comprises the steps that a target feature map corresponding to a to-be-detected image is acquired, the target feature map comprises N anchor point frames, N is an integer larger than 1, the confidence score of each anchor point frame is calculated, the confidence score is used for indicating that the anchor point frame contains a probability value of a target object, and the probability value of each anchor point frame contains the target object; m target anchor point frames are determined from the N anchor point frames according to the confidence score, M is an integer larger than 1 and smaller than N, key point calculation is carried out on each target anchor point frame in the M target anchor point frames to obtain M key point coordinates, and the position of a parking space in the to-be-detected image is determined according to the M key point coordinates. And the position of the parking space in the to-be-detected image is pushed to the target terminal device.

Description

technical field [0001] The embodiments of the present application relate to the technical field of intelligent transportation, and in particular to a detection method, device, equipment and storage medium of a parking space. Background technique [0002] With the intensification of urbanization and the improvement of people's living standards, there are more and more cars in big cities, and most urban areas have a large flow of people. It is very difficult to find parking spaces during peak hours. The problem of difficult parking is becoming more and more prominent. Vehicle users urgently need to accurately know the status information of free parking spaces in the parking lot to improve parking efficiency. [0003] At present, the detection method of free parking space is usually to detect the entry point of the parking space based on deep learning, and then combine the artificial prior knowledge to estimate the depth of the parking space, and then generate the location inf...

AI Technical Summary

Problems solved by technology

[0004] However, the depth estimation of the parking space not only depends on the correct prediction of the entry point, if the predicted entry point does not fit the entry point in the actual image well, it is easy to cause the position of the detected parking space to be relatively poor. Large angle deviation, and it also depends on artificial prior knowledge. The artificially set rules cannot cover various parking spaces, and it is easy to increase the detection error of parking spaces, resulting in low accuracy in detecting parking spaces, which leads to parking efficiency. reduce

Images