Research And Application Of 3D Model Retrieval Algorithms For Aircraft Parts

With the rapid development of aviation manufacturing industry,the demand for aviation products is growing rapidly.At the same time,higher requirements are put forward for the quality of aviation products.Spraying is an important process in the production process of aviation products,and its quality seriously affects the service performance and life of products.At present,the spraying process is mainly based on the traditional manual and semi-automatic production mode,and does not make accurate spraying trajectory for the three-dimensional shape of parts,and spraying accuracy is poor,efficiency is low,waste of materials and other issues.In this paper,computer technology is introduced into aeronautical spraying production line.In the process of reconstructing the three-dimensional model of Aeronautical parts and obtaining the pre-designed spraying trajectory,the three-dimensional model retrieval algorithm is used to independently identify the aeronautical parts,so as to achieve high-precision and high-efficiency aeronautical parts spraying.Aiming at the traditional three-dimensional model retrieval technology,the algorithm is further optimized.According to the structural characteristics of the three-dimensional model of aircraft parts,an improved algorithm based on D2 shape distribution feature extraction is proposed to make it more suitable for the retrieval of the three-dimensional model of aircraft parts.D2 shape distribution feature extraction algorithm can achieve a good feature extraction and accurate matching with the model by calculating the distance between two random points on the surface of three-dimensional model of aircraft parts.Compared with other retrieval algorithms,this algorithm is simpler and faster.It can extract fewer features from the surface of three-dimensional model of aircraft parts for retrieval.It is suitable for retrieval of three-dimensional model of aircraft parts with fewer parts in different categories,fewer parts in each large category and without holes and holes.By introducing a model segmentation method which has a good description of local features,and based on the global feature extraction algorithm based on D2 shape distribution,combined with their respective advantages.In this paper,a 3D model retrieval algorithm based on D2 shape distribution and model segmentation is proposed.This algorithm combines the global and local features of the 3D model,and achieves more accurate retrieval of the 3D model of aircraft parts.The experimentalresults show that the retrieval accuracy of this method is higher than that of the global feature extraction algorithm.Under the same recall rate,higher accuracy can be obtained,which can be applied to the retrieval of the same type of parts and the retrieval of parts with little difference in overall details.Based on the above two retrieval algorithms,this paper designs and implements a three-dimensional model retrieval system for aircraft parts based on B/S architecture.The system has good expansibility,including four modules: off-line computing module,online retrieval module,three-dimensional model management module and information management module,and achieves high-speed and accurate retrieval function for three-dimensional model of aircraft parts.