Drawing Computerization and Intelligent Database Building Algorithm

In engineering, manufacturing, electronic industry, transportation and other industries, a large number of paper drawings have been accumulated. These paper drawings are an important resource accumulation of the companies. Paper drawings have some disadvantages, such as inconvenient to save, unable to establish database to realize resource reuse and so on. Therefore, this paper proposes a new algorithm of unattended drawing computerization and database intelligent establishment. Firstly, the position of the key text area is located according to the deep learning object detection algorithm, then the text lines in the text area are detected and the text is recognized. Finally, the database is established according to the recognition results. The experimental results show that the average accuracy of the proposed algorithm is 98.6%. Compared with the existing drawing retrieval algorithm, the algorithm uses deep learning object detection algorithm to initially locate the text area. It can further improve the accuracy of drawing information extraction, and realize the unattended paper drawing computerization and intelligent establishment of database.