Shijie Cai

An object-oriented progressive-simplification-based vectorization system for engineering drawings: model, algorithm, and performance

Existing vectorization systems for engineering drawings usually take a two-phase workflow: convert a raster image to raw vectors and recognize graphic objects from the raw vectors. The first phase usually separates aground truth graphic object that intersects or touches other graphic objects into several parts, thus, the second phase faces the difficulty of searching for and merging raw vectors belonging to the same object. These operations slow down vectorization and degrade the recognition quality. Imitating the way humans read engineering drawings, we propose an efficient one-phase object-oriented vectorization model that recognizes each class of graphic objects from their natural characteristics. Each ground truth graphic object is recognized directly in its entirety at the pixel level. The raster image is progressively simplified by erasing recognized graphic objects to eliminate their interference with subsequent recognition. To evaluate the performance of the proposed model, we present experimental results on real-life drawings and quantitative analysis using third party protocols. The evaluation results show significant improvement in speed and recognition rate.

A new recognition model for electronic architectural drawings

Current methods for recognition and interpretation of architectural drawings are limited to either low-level analysis of paper drawings or interpretation of electronic drawings that depicts only high-level design entities. In this paper, we propose a Self-Incremental Axis-Net-based Hierarchical Recognition (SINEHIR) model for automatic recognition and interpretation of real-life complex electronic construction structural drawings. We design and implement a series of integrated algorithms for recognizing dimensions, coordinate systems and structural components. We tested our approach on more than 200 real-life drawings. The results show that the average recognition rate of structural components is about 90%, and the computation time is significantly shorter than manual estimation time.

Automatic analysis and integration of architectural drawings

Recognition and integration of 2D architectural drawings provide a sound basis for automatically evaluating building designs, simulating safety, estimating construction cost or planning construction sequences. To accomplish these targets, difficulties come from (1) an architectural project is usually composed of a series of related drawings, (2) 3D information of structural objects may be expressed in 2D drawings, annotations, tables, or the composites of above expressions, and (3) a large number of disturbing graphical primitives in architectural drawings complicate the recognition processes. In this paper, we propose new methods to recognize typical structural objects and architectural symbols. Then the recognized results on the same floor and drawings of different floors will be integrated automatically for accurate 3D reconstruction.

Line net global vectorization: an algorithm and its performance evaluation

In this paper, an efficient global algorithm for vectorizing line drawings is presented. It first extracts a seed segment of a graphic entity from a raster image to obtain its direction and width, then tracks the pixels under the guidance of the direction so that the tracking can track through functions and is not affected by noise and degradation of image quality. Thus, an entity will be vectorized in one step without postprocessing. The relations among lines are also used to realize the continuous vectorization of a line net. The speed and quality of vectorization are greatly improved with this algorithm. The performance evaluation is carried out both by theoretical analysis and by experiments. Comparisons with other vectorization algorithms are also made.

3D Reconstruction of Detailed Buildings from Architectural Drawings

AbstractDue to the difficulties in designing and modifying a complex 3D architectural building in 3D space, 2D drawings are often a more effective means for design. However, for various applications such as quantity surveying and visualization, it is necessary to convert 2D architectural drawings to 3D models. In this paper, we propose a new method for accurate 3D reconstruction from real-life architectural drawings. The method integrates and normalizes architectural information dispersed in multiple drawings and tables under the guidance of semantics and prior domain knowledge. The reconstructed detailed 3D building can be used for quantity surveying, construction, and 4D modeling.

A Novel Knowledge-Based System for Interpreting Complex Engineering Drawings: Theory, Representation, and Implementation

We present a novel knowledge-based system to automatically convert real-life engineering drawings to content-oriented high-level descriptions. The proposed method essentially turns the complex interpretation process into two parts: knowledge representation and knowledge-based interpretation. We propose a new hierarchical descriptor-based knowledge representation method to organize the various types of engineering objects and their complex high-level relations. The descriptors are defined using an Extended Backus Naur Form (EBNF), facilitating modification and maintenance. When interpreting a set of related engineering drawings, the knowledge-based interpretation system first constructs an EBNF-tree from the knowledge representation file, then searches for potential engineering objects guided by a depth-first order of the nodes in the EBNF-tree. Experimental results and comparisons with other interpretation systems demonstrate that our knowledge-based system is accurate and robust for high-level interpretation of complex real-life engineering projects.

Dimension recognition and geometry reconstruction in vectorization of engineering drawings

This paper presents a novel approach for recognizing and interpreting dimensions in engineering drawings. It starts by detecting potential dimension frames, each comprising only the line and text components of a dimension, then verifies them by detecting the dimension symbols. By removing the prerequisite of symbol recognition from detection of dimension sets, our method is capable of handling low quality drawings. We also propose a reconstruction algorithm for rebuilding the drawing entities based on the recognized dimension annotations. A coordinate grid structure is introduced to represent and analyze two-dimensional spatial constraints between entities; this simplifies and unifies the process of rectifying deviations of entity dimensions induced during scanning and vectorization.

A Knowledge-Aided Line Network Oriented Vectorisation Method for Engineering Drawings

Abstract:Vectorisation is the foundation in recognising engineering components from paper form drawings. Due to the complexity of problems and the difficulty of techniques, the vectorisation method relying merely on the image itself cannot get satisfactory results. It is now widely agreed that the knowledge must be applied more or less to aid the vectorisation. The capability of the vectorisation method itself should also be thus improved. This paper analyses the problems of existing vectorisation methods, introduces the complete concept of global vectorisation, and proposes a whole new line network oriented global vectorisation method. This method uses global information to vectorise a line in one step, and carries out the global vectorisation of line networks. Therefore, the problem of separating one line is solved, and a complex analysis of crossings is avoided. The performance of vectorisation is improved clearly. Furthermore, it can vectorise lines in any orientations well, and can vectorise a dashed line in one step. Aided by the related knowledge, local details of vectorisation are refined. A performance evaluation compared with other vectorisation methods is also included.

A vectorization system for architecture engineering drawings

This paper presents a vectorization system for architecture engineering drawings. The system employs the line-symbol-text vectorization workflow to recognize graphic objects in the order of increasing characteristic complexity and progressively simplify the drawing image by removing recognized objects from it. Various recognition algorithms for basic graphic types have been developed and efficient interactive recognition methods are proposed as complements to automatic processing. Based on dimension recognition and analysis, the system reconstructs the literal dimension for vectorization results, which yields optimized vector data for CAD applications.

Raster to vector conversion of construction engineering drawings

Abstract A novel method for vectorizing construction engineering drawing images is presented. The method consists of two algorithms: a seed-segment based line-network vectorization algorithm and a knowledge-supported vectorization algorithm designed specially for construction engineering drawings. The seed-segment based algorithm converts a line wholly and a line network globally so that both the quality and speed are satisfied. The knowledge supported vectorization algorithm takes full use of the features and rules of construction engineering drawings to solve the difficulties of vectorization caused by noises and degradations. The performance of this method is discussed at the end of this paper.