Splicing method for tire mold images

Abstract

The invention discloses a splicing method for tire mold images, which comprises the steps of inputting a set of photographed tire mold images, extracting the external boundary of the photographed tire mold images, fitting the external boundary to be arc-shaped, straightening the arc-shaped external boundary through the polar transformation process, conducting the pretreatment and the threshold segmentation, conducting the registration based on the gradient and phase-related method to obtain the overlapped part of two adjacent images, fusing by adopting a fusion method based on the combination of the optimal convergence and the weighted gradient fusion, splicing all adjacent images to form a first image based on the fusing result, determining the fore-and-aft overlapped part of the first image based on the registration algorithm, processing the first image in combination with the above fusion method to obtain a second image, and restoring the second image into the image of a circular tire mold through the polar coordinate inverse-transformation process. The method is good in splicing effect, simple, convenient and practical, which can be widely applied to the image processing field.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a splicing method of tire mold images. Background technique [0002] With the development of science and technology, image mosaic has become a very important technology in the field of image information processing, and is widely used in digital video, motion analysis, virtual reality technology, medical image analysis and remote sensing image processing and other fields. Image stitching refers to the stitching of multiple concatenated images into a high-resolution panorama according to the overlapping parts of the images. At present, the splicing technology based on feature points is relatively mature, and it is suitable for images with many feature points. The calculation amount of this algorithm is small and the registration accuracy is high, but it often needs to select the initial matching points manually, which greatly reduces the accuracy of the algorithm. speed and scope o...

AI Technical Summary

Problems solved by technology

At present, the splicing technology based on feature points is relatively mature, and it is suitable for images with many feature points. The calculation amount of this algorithm is small and the registration accuracy is high, but it often needs to select the initial matching points manually, which greatly reduces the accuracy of the algorithm. The speed and scope of

The stitching algorithm based on phase correlation is also a commonly used method, which has the advantages of high registration accuracy and fast speed, but it is not suitable for stitching images with rotation relationship

[0003] Due to the limitation of the characteristics of the tire mold image itself, it is difficult to register the tire mold image when splicing: some overlapping areas of the pictures have only diagonal lines and no text, and some overlapping areas of the pictures have simple lines, which may easily cause difficulty in matching feature points or feature points Insufficient; the uneven illumination of the image will also increase the difficulty of preprocessing

Traditional stitching algorithms generally only target two or a few images, but for tire mold images that require multiple images (generally 30 to 40 images) to be stitched, the stitching effect of traditional algorithms is not ideal

In addition, due to the inability to meet the speed requirements of industrial production and the inability to solve the problem of image distortion encountered during shooting, traditional algorithms are not suitable for the stitching of tire mold images

Therefore, currently in engineering applications, there are few mosaic algorithms for tire mold images

Images