Python 如何获取用于校准摄像机的摄像机矩阵

我使用的是acA4112-30uc-Basler ace相机。我试图消除失真，但为了做到这一点，我必须获得“mtx”参数，我不知道如何获得该参数。

使用OPENCV进行相机校准的理论如下：

首先你需要修理照相机。然后打印棋盘

A部分：拍摄图像并进行计算

印刷后，把它放在一本书上。然后使用以下代码通过相机拍摄一些图像：

import numpy as np
import cv2

objp = np.zeros((6 * 7, 3), np.float32)
objp[ : , : 2] = np.mgrid[0 : 7, 0 : 6].T.reshape(-1, 2)
objpoints = []
imgpoints = []

criteria = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_MAX_ITER, 30, 0.001)

cam = cv2.VideoCapture(0)
(w, h) = (int(cam.get(4)), int(cam.get(3)))

while(True):
    _ , frame = cam.read()
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    ret, corners = cv2.findChessboardCorners(gray, (7, 6), None)

    if ret == True:
        objpoints.append(objp)
        corners = cv2.cornerSubPix(gray, corners, (11, 11), (-1, -1), criteria)
        imgpoints.append(corners)

        cv2.drawChessboardCorners(frame, (7, 6), corners, ret)
        cv2.imshow('Find Chessboard', frame)
        cv2.waitKey(0)
    cv2.imshow('Find Chessboard', frame)
    print "Number of chess boards find:", len(imgpoints)        
    if cv2.waitKey(1) == 27:
        break

ret, oldMtx, coef, rvecs, tvecs = cv2.calibrateCamera(objpoints, imgpoints,
                                                      gray.shape[: : -1], None, None)
newMtx, roi = cv2.getOptimalNewCameraMatrix(oldMtx, coef, (w, h), 1, (w, h))

print "Original Camera Matrix:\n", oldMtx
print "Optimal Camera Matrix:\n", newMtx

np.save("Original camera matrix", oldMtx)
np.save("Distortion coefficients", coef)
np.save("Optimal camera matrix", newMtx)

cam.release()
cv2.destroyAllWindows()

在这段代码中，您打开相机并将打印的棋盘放在多个位置，对用于矩阵计算的棋盘进行编码检测。每次代码检测棋盘显示图像并打印检测到的图像数。在一些图像之后，可以按Esc键完成计算

第B部分：使用矩阵校正摄像机输出

您可以将此代码用于测试：

import numpy as np
import cv2

oldMtx = np.load("Original camera matrix.npy")
coef = np.load("Distortion coefficients.npy")
newMtx = np.load("Optimal camera matrix.npy")

cam = cv2.VideoCapture(0)
(w, h) = (int(cam.get(4)), int(cam.get(3)))

while(True):
    _ , frame = cam.read()

    undis = cv2.undistort(frame, oldMtx, coef, newMtx)

    cv2.imshow("Original vs Undistortion", np.hstack([frame, undis]))
    key = cv2.waitKey(1)
    if key == 27:
        break
cam.release()
cv2.destroyAllWindows()

祝你好运