Design of Digital Museum System based on AR and ORB Optimization Algorithm

Aiming at the problems of high cost and narrow popularization of existing augmented reality technology, which is not suitable for most museum exhibitions, the study introduces the feature point detection idea of binary robust invariant scalable key point algorithm to improve the matching recognition algorithm based on natural features of images. The study found that the matching time of the improved matching recognition algorithm is below 1s and the matching accuracy rate is above 90%. The augmented reality system based on the improved matching recognition algorithm can successfully recognize the exhibits with the recognition distance less than 80cm and the recognition characteristics blocked by 50%, and can efficiently establish the corresponding three-dimensional model of the exhibits in the digital museum. To sum up, the improved matching recognition algorithm proposed in this paper is robust and has certain application value in the exhibition of digital museum exhibits.