Panoramic parking assistance system and its wide-angle image debugging method and device

Abstract

The invention provides a wide-angle image debugging method and device of a panoramic parking assistance system. The wide-angle image debugging method comprises the steps that the world coordinates of debugging template feature points are acquired; the image data of the debugging template feature points acquired by a wide-angle camera are acquired, wherein the image data at least include the theoretical coordinate values of the debugging template feature points in the acquired images; an observation model between the image observation values of the feature points and internal and external parameter sets is established according to the acquired image data of the debugging template feature points; the error between the current image observation values of the feature points and the theoretical coordinate values is calculated; and the error is iteratively optimized so that the optimized internal and external parameter sets are obtained. The feature point observation model is established through the internal and external parameter sets so that the internal and external parameters can be optimized in optimization iteration, and the debugging error caused by the input error of the internal parameters in the debugging process can be avoided.

Description

technical field [0001] The invention relates to the field of video image processing, in particular to a panoramic parking assistance system and a wide-angle image debugging method and device thereof. Background technique [0002] The panoramic parking assist system, also known as the 360-degree panoramic assist system, mainly uses ultra-wide-angle cameras installed in the front, rear, left, and right directions of the vehicle body to collect images, and after de-distorting and splicing the images collected by the cameras, it shows that the effect is quite A panoramic view of the car's surroundings from the roof. [0003] The stitching of panoramas requires two parameters: the internal and external parameters of the camera. The internal parameters reflect the characteristics of the camera itself. The distortion of the camera itself can be removed through the adjustment of the internal parameters, which mainly include focal length, center point, and distortion parameters. (...

AI Technical Summary

Problems solved by technology

However, because the template used in the current step of debugging is too simple, it is necessary to use the keyboard to interact with the screen, and the on-site debugging personnel to complete the calibration of the feature points, which takes a long time and is not conducive to automated mass production

Images