A study on the background noise error of autocollimators based on Retinex theory

Subject of study . Autocollimation measurement systems are the main means to realize precision angle measurement, but multifaceted error sources limit the further improvement of autocollimator angle measurement accuracy. The purpose of the work. This paper focuses on a kind of reflection noise induced from the non-working optical structure surface inside the autocollimator. This noise source can lead to errors in the detection of the target image by the image sensor. Method. Combined with the principle of Hough transform image recognition, we propose to utilize the Retinex theory to compensate the reflection noise inside the autocollimator. Main results. It is experimentally verified that this compensation algorithm can effectively suppress the influence of noise on imaging detection, and the measurement accuracy of the autocollimator around the X -axis and Y -axis can be improved from 4.29 ′ ′ and 3.87 ′ ′ to 3.59 ′ ′ and 3.15 ′ ′ . The system measurement stability performance is also improved, about 11.99% and 15.75% in the X -axis and Y -axis directions, as well. Practical significance. Therefore, this algorithmic compensation method can effectively enhance the measurement performance of the autocollimator.