Image stitching method in panoramic surround-view system

Abstract

The invention relates to an image stitching method in a panoramic surround-view system, and relates to the field of computer vision. The method includes the steps of 1) correcting the fisheye image distortion by combining a spaced line training model and a SVM algorithm; 2) establishing an image overlook transformation look-up table to achieve a quick overlook transformation and obtaining a bird's-eye view; 3) generating a panoramic image stitching mapping table, and quickly obtaining a panoramic surround-view bird's-eye view by looking up the panoramic image stitching mapping table; and 4) solving the brightness difference between every two stitched images by using an improved brightness tuning processing algorithm, and then further eliminating stitching seams by using an image fusion method of a weighted average method. The method of the invention effectively reduces the system overhead, quickly achieves the seamless image stitching, and obtains the panoramic surround-view bird's-eye view.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to an image splicing method in a panoramic view system. Background technique [0002] With the rapid development of the country's economy, the total number of cars continues to increase, and the traffic environment becomes increasingly congested, causing cars to often drive in narrow environments. Cars park in a narrow parking lot through narrow roads or traffic flow. Due to the limited vision of the driver, it is easy to collide and cause unnecessary losses. With the continuous reduction of the cost of digital processors, cameras and other equipment, the car auxiliary system based on the panoramic view system will become the mainstream of the future automotive visual assistance system, and the important technical means of the panoramic view system is the stitching of the panoramic view images. The panorama image is obtained through panoramic surround view image stitching and output ...

AI Technical Summary

Problems solved by technology

The fisheye lens can obtain a wide range of viewing angles and a complete spherical image, but the complex optical structure of the fisheye lens causes serious distortion in the fisheye image, so image distortion correction must be performed before image stitching

Use the method of spatial coordinate system transformation to realize perspective transformation. This method is simple and effective. It can be calculated as long as the camera installation parameters (camera position, angle and other installation parameters) are determined. does not get the correct top view

Although the region-based image stitching method and the feature-based image stitching method can complete the image stitching, both stitching methods require a certain range of overlapping areas between the stitched images, and the images should not be too distorted.

However, there is a certain degree of distortion in the image of the overlooking transformation and the image difference calculation in the process of panoramic image stitching. Therefore, the above two stitching methods cannot achieve good results in image stitching, and even the panoramic image cannot be stitched together.

Images