66 lines
1.7 KiB
Python
66 lines
1.7 KiB
Python
'''
|
|
完成相机外参矩阵的标定
|
|
|
|
'''
|
|
|
|
|
|
import cv2
|
|
import numpy as np
|
|
|
|
|
|
# 标定板参数
|
|
pattern_size = (11, 8) # 标定板格点数
|
|
square_size = 0.020 # 每个格子的边长（单位：米）
|
|
|
|
# 准备标定板在实际世界中的坐标
|
|
objp = np.zeros((np.prod(pattern_size), 3), dtype=np.float32)
|
|
objp[:, :2] = np.indices(pattern_size).T.reshape(-1, 2)
|
|
objp *= square_size
|
|
|
|
# 存储实际世界中的点坐标和图像中的点坐标
|
|
obj_points = []
|
|
img_points = []
|
|
|
|
# 读取标定板图像
|
|
images = ["1724898532.521401.jpg"] # 替换为实际的图像文件
|
|
|
|
for fname in images:
|
|
img = cv2.imread(fname)
|
|
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
|
|
|
|
# 找到棋盘格角点
|
|
ret, corners = cv2.findChessboardCorners(gray, pattern_size)
|
|
if ret:
|
|
obj_points.append(objp)
|
|
img_points.append(corners)
|
|
|
|
# 可视化角点
|
|
cv2.drawChessboardCorners(img, pattern_size, corners, ret)
|
|
# cv2.imshow('Corners', img)
|
|
cv2.waitKey(500)
|
|
|
|
cv2.destroyAllWindows()
|
|
|
|
# 相机内参和畸变系数（通常先通过标定获取）
|
|
camera_matrix = np.array([[1461.09955384437, 0, 963.862411991572],
|
|
[0, 1461.89533915958, 545.053909593624],
|
|
[0, 0, 1]])
|
|
dist_coeffs = np.array([0.123539164198816, -0.224109122219162, 0, 0, 0])
|
|
|
|
# 使用solvePnP计算外参
|
|
retval, rvec, tvec = cv2.solvePnP(obj_points[0], img_points[0], camera_matrix, dist_coeffs)
|
|
|
|
# 将旋转向量转换为旋转矩阵
|
|
R, _ = cv2.Rodrigues(rvec)
|
|
|
|
# 平移向量
|
|
T = tvec
|
|
|
|
# 打印旋转矩阵和平移向量
|
|
print("旋转矩阵 R:\n", R)
|
|
print("平移向量 T:\n", T)
|
|
|
|
|
|
# 保存相机参数到文件
|
|
np.savez("camera_params.npz", camera_matrix=camera_matrix, dist_coeffs=dist_coeffs, R=R, T=T)
|