Growth-based checkerboard corner detection method

0 Introduction

Opencv has a function for corner detection, but it has the following three disadvantages:

1. The checkerboard size needs to be specified in advance.
2. Robust angle difference.If the chessboard is inclined at a large angle, the detection will fail.
3. It cannot handle the situation that a picture contains multiple chessboards.

The checkerboard corner points based on the growth method have the following characteristics:

Advantages:
1. No need to specify the number of checkerboards in advance.
2. Good robustness.Because it is a growth-based algorithm, if there is interference, it will bypass the interference and grow the largest chessboard.
3. It can detect the situation that a picture contains multiple chessboards.

Disadvantages:
1. Large amount of calculation.Mainly focus on the chessboard growing part.

Rendering:

1 Corner Detection

1.1 generate kernel