Research On Content-based Vector Engineering Drawings Retrieval

The reuse of electronic drawings is an effective way to improve the efficiency of CAD software. How to speedily, accurately search out the required drawings from the database is especially important to the user. Nowadays, the query of drawings is performed based on keywords, such as project, designer, drawing number, and so on. The keyword-based query has the merit of quick speed, easy technology realization and use. However, the retrieval way based on keywords has lots of limitations which are difficult to choose keywords and completely describe engineering drawings only by keywords.In view of the problems in keyword-based retrieval and combined with the characteristics of the vector engineering drawings, this paper presents a system framework of the content-based vector engineering drawings retrieval technology and describes the involved key technologies. Then it introduces the pretreatment process of engineering drawings and the related techniques, including coordinates normalization, the decomposition of engineering drawings and so on. For the engineering drawings that have been decomposed, according to the extracted topological features and geometry features of engineering drawings, the paper proposes a similarity calculation method based on the topological tree, which can improve the recall ratio of vector engineering drawings retrieval under the premise of ensuring the precision ratio. Aimed at the problem of the low efficiency of the single indexing of large drawings database, the indexing technique is researched to this paper and a two-tier indexing method is proposed. The first layer is a simple classification index; and the second layer is a multi-dimensional indexing. In the second layer we design a two-dimensional KD-tree indexing method. Based on the dimension reduction in high-dimensional vector, the method realizes the efficient search of the mass engineering drawings database. Finally, the paper gives the technology architecture of the content-based vector engineering drawings retrieval and completes the design and development of the main functions. We introduce the application into the product data management system in certain enterprise.