Two-dimensional asynchronous coding comparison method

Abstract

The present invention discloses a two-dimensional asynchronous coding comparison method, which not only considers the difference of rotation of pixels, but also the fault tolerance of segmentation errors of the inner and outer circles of the iris is taken into account, so that the two-dimensional asynchronous comparison of the iris pixels in the angular direction and in the radial direction is realized. Compared with the conventional one-dimensional pixel comparison method (i.e., only considering movement correction in the iris angular direction), segmentation errors in the iris pixel processis considered in the method disclosed by the present invention; compared with the synchronized two-dimensional pixel comparison (i.e., the overall movement correction in the radial direction and in the angular direction), the comparison method disclosed by the present invention is more flexible; and according to the method disclosed by the present invention, not only the synchronous two-dimensional comparison method is covered, but also misalignment and overlapping comparison of the two-dimensional pixel within a reasonable range is allowed, so that the method can better adapt to the center shift and radius errors occur in the iris segmentation process, and has advantages of a small calculation amount and a fast detection speed.

Description

Technical field [0001] The invention relates to the technical field of iris recognition, in particular to a two-dimensional asynchronous coding comparison method. Background technique [0002] The texture feature of the random distribution of iris is a natural and high-confidence identification of human beings. Therefore, iris recognition becomes one of the important means of identity verification due to the development trend of technology and market. [0003] The iris recognition process generally includes iris segmentation, transformation of iris space coordinates to polar coordinates, iris texture enhancement, iris pixels, and pixel comparison. The main content of the present invention is the pixel comparison link in the iris recognition process. [0004] The iris is an organ used to adjust the size of the pupil. It is circular when viewed from the front. The inner and outer circles of the iris refer to the outer edge of the pupil and the outer edge of the iris respectively; the ...

AI Technical Summary

Problems solved by technology

Since the pixels strictly depend on the segmentation results of the inner and outer circles, the seemingly reasonable segmentation error may lead to misjudgment in some cases

Still using the example of the elliptical pupil mentioned above, the result of segmentation may be biased towards the circle with the largest number of fitting points or an average circle. It is reasonable to consider both cases from the perspective of fitting, but in space coordinates In the process of converting to polar coordinates, the expanded image may be very different, and this difference will also lead to differences in pixels

In this case, it is obviously unreasonable to only consider the difference in rotation and ignore the segmentation error

Images