| 摘要 |
The present invention relates to the technical field of image processing, particularly to a highly robust mark point decoding method and system. The decoding method comprises the following steps: step A, estimating a homography matrix and converting perspective projection images of mark points into orthographic projection images; step B, traversing coding segments of the mark points images in a polar coordinate system to obtain a pixel value in a cartesian coordinate system corresponding to each point, and determining the length of each coding segment and the code value thereof in a binary coding sequence, so as to determine the number of code value bits of each coding segment in the binary coding sequence to form a binary coding sequence; step C, performing cyclic shifting on the binary coding sequence, and converting each cyclic-shifted sequence into a decimal coded value, and marking the minimum decimal coded value as a coded value of the mark point. The present invention can better avoid errors in determination of coded feature areas in mark points caused by a photographing angle, camera resolution and noise. |
