Algorithm 2: Pseudocode of the perturbation simulation experiment.

For $i=1:5$ ($i$ is the number of groups in the experiment)     For $j=1:5$ ($j$ is the number of images taken in each group)       For $k=1:10$ ($k$ is the duration over which Gaussian noise was added)         (I) Obtain image $I_j$, and record values of the inclinometer $\left({\theta }_j,{\varphi }_j\right)$.         (II) Detect plumb lines in the image $I_j$ using the Hough line detection method.         (III) Add Gaussian noise to the endpoints of detected lines $\left(x_n^{j,k},y_n^{j,k}\right)$.       End     End      (IV) Calculate the inclinometer assembly error $R_{ic}^i$, and decompose the matrix into Euler angles.   End   (V) Calculate the average of the Euler angles as the final assembly error for each group.   (VI) Calculate the mean and the standard deviation of the Euler angle.