To achieve dynamic attitude monitoring of shipboard equipment and overcome the limited generality of traditional single-view methods, this paper proposes a multi-feature-point, multi-view attitude measurement approach.The method is designed as a redundant supplement to existing attitude measurement systems, providing additional attitude information under complex operating conditions.An improved MobileNet-v2 network incorporating squeeze-and-excitation modules is employed for feature point extraction from equipment images, achieving a detection accuracy of 95.4% on the test set.Based on multi-view observations and known equipment geometry, the three-dimensional coordinates of feature points in the camera coordinate system are reconstructed, and the equipment attitude is estimated
Wang et al. (Thu,) studied this question.