Vehicle parking distance measurement method and device based on deep learning, and electronic equipment

Abstract

The invention discloses a vehicle parking distance measurement method and device based on deep learning and electronic equipment, and the method comprises the steps: determining a camera distortion coefficient, and determining a homography transformation matrix based on an installed camera according to the image coordinates and actual physical coordinates of all identification points in a preset marker; when the vehicle is in a side parking state, acquiring an image acquired by the camera, detecting a white line of a roadside in the acquired image by adopting a semantic segmentation algorithm based on deep learning, and outputting a binary image; removing lens distortion in the binary image based on the distortion coefficient and a distortion removing algorithm, and obtaining a distortion-removed image; and performing back projection transformation through a homography transformation matrix to obtain an image with uniform resolution, and measuring the distance between the vehicle and the roadside white line. According to the invention, the position of the roadside white line can be accurately detected in different scenes, the detection accuracy of the position of the white line and the actual application scene are improved, lens distortion in camera imaging is removed, and the distance between the vehicle and the white line is accurately measured.

Description

technical field [0001] The invention relates to the fields of computer vision and advanced driving assistance, in particular to a method, device and electronic equipment for vehicle parking distance measurement based on deep learning. Background technique [0002] In the third test of the driving test subject, the side parking module of the road test instrument measures the distance between the vehicle and the white line on the roadside by taking pictures with the camera installed on the vehicle, so as to judge whether the student has reached the distance between the vehicle and the roadside required by the test. The white line is within the range of 30cm to 50cm. [0003] However, most of the existing road test instrument products use traditional image processing methods, such as threshold segmentation, Hough transform and other simple methods to detect white lines from images. These simple methods cannot cope with the complex and changeable lighting, weather and road condi...

AI Technical Summary

Problems solved by technology

[0005] The main purpose of the present invention is to provide a vehicle parking distance measurement method and device based on deep learning to solve the problem that the existing road test instrument cannot correctly detect the white line on the roadside and cannot measure the accurate distance between the vehicle and the white line

Images