A processor for calibrating an automotive look-around system

Abstract

The present invention relates to a processor for calibrating an automotive look-around system, the automobile look-around system comprises a plurality of fish-eye cameras, the processor is configuredto execute the computer program steps of obtaining the rotation matrix Rt of the cameras, when the automobile stops, obtaining the observation vectors v1, v2 and v3 of the camera to three vertices ofa known marker under a camera coordinate system, and calculating a normal vector z of an equilateral triangle under the camera coordinate system according to a triangle side length and a spatial geometric relationship; when the vehicle moves along the X axis of the vehicle coordinate system, calculating the x axis unit vector of the camera coordinate system; calculating the rotation matrix of thecamera. Compared with the prior art, the process adopted by the processor of the invention makes full use of the potential prior information, the calibration scheme is simple, the calibration space required is small, the calibration precision is high, the anti-interference is stronger, and the requirement on the skill of the operator is greatly reduced to the extent that the driver can do it by himself.

Description

technical field [0001] The invention relates to a processor with a computer program, in particular to a processor for calibrating a car surround view system. Background technique [0002] With the increase of car ownership and the increasingly complex traffic environment, safe parking has become a difficult problem for drivers. In order to reduce the potential safety risks in the parking process, various auxiliary parking systems have emerged, such as rear-view reversing camera, reversing radar and other systems. These systems can reduce the driver’s parking burden to a certain extent, but there are still unintuitive And can not completely eliminate the defects of blind spots. The 360-degree surround view system uses multiple cameras to cover the 360-degree area around the vehicle body to eliminate visual blind spots. At the same time, the 360-degree surround view system seamlessly stitches the images around the vehicle body into a bird's-eye view and presents it to the dri...

AI Technical Summary

Problems solved by technology

[0003] Although the value of the 360-degree surround view system in the entire vehicle is not high, it plays an important role in factory calibration. The reason is that the current calibration scheme still requires a lot of manpower, which increases the Proportion to the cost of the entire vehicle

[0004] Due to the assembly deviation of the same model and the installation deviation of the cameras around the body, if a set of fixed parameters is used for 360-degree surround view system image stitching, these deviations will affect the image stitching effect, so offline calibration must be performed to eliminate them The impact of these deviations

The traditional factory calibration scheme has the following problems: (1) It has high requirements for the calibration environment, such as the need to park the car at a fixed position, and requires high precision, which increases the technical difficulty of workers' operations and takes up more manpower and time costs ; (2) This method is difficult to apply to 4S store calibration, and it is even more difficult for drivers to use it personally, which indirectly increases the potential car purchase cost

[0006] The current online calibration scheme based on lane line detection has disadvantages such as inaccurate length and width of lane lines, misalignment of adjacent lane lines, difficulty in calibration of left and right cameras, and potential safety hazards in the calibration process (usually on expressways).

Images