Distort fisheye lens images with asymmetric_grid

Python
,
1

I want to correct the distortion image in the fisheye lens. I need to find the camera calibration (K) and distortion factor (D) values, but the K value is calculated and the D value is all zero and not calculated. I tried adding the data and changed the initial D value to a very small value, but the D value keeps coming out to zero.
Is there any other way to solve it?

Picture11
Picture113261×2436 1.23 MB

2

and how did you do that? MRE please or something like it.

3

Hello, here is my code :

import cv2
import numpy as np
import glob

patternSize = (3, 13) 

objpoints = []
imgpoints = []  

objp = np.zeros((1, patternSize[0] * patternSize[1], 3), np.float32)
objp[0,:,:2] = np.indices(patternSize).T.reshape(-1, 2)


img = cv2.imread('./img_0221/grid/0221095408.jpg')
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
ret, centers = cv2.findCirclesGrid(img, patternSize, None, cv2.CALIB_CB_ASYMMETRIC_GRID+cv2.CALIB_CB_CLUSTERING)
    
if ret:
     imgpoints.append(centers)
     objpoints.append(objp)

K = np.zeros((3, 3))
D = np.zeros((4, 1))
retval, K, D, rvecs, tvecs = cv2.fisheye.calibrate(
    objpoints,
    imgpoints,
    gray.shape[::-1],
    K,
    D,
    None,
    None,
    cv2.fisheye.CALIB_RECOMPUTE_EXTRINSIC,
    (cv2.TERM_CRITERIA_EPS + cv2.TERM_CRITERIA_MAX_ITER, 30, 1e-6)
)

print("K:", K)
print("D:", D)
4

calibration doesn’t work on just a single picture.

5

I tested with single image and 50 to 80 images. But, i had same problem.