P.M. Devaux’s research while affiliated with Pennsylvania State University and other places

. Converting paper-based engineering drawings into CAD model files is a tedious process. Therefore, automating the conversion of such drawings represents tremendous time and labor savings. We present a complete system which interprets such 2D paper-based engineering drawings, and outputs 3D models that can be displayed as wireframes. The system performs the detection of dimension sets, the extraction of object lines, and the assembly of 3D objects from the extracted object lines. A knowledge-based method is used to remove dimension sets and text from ANSI engineering drawings, a graphics recognition procedure is used to extract complete object lines, and an evidential rule-based method is utilized to identify view relationships. While these methods are the subject of several of our previous papers, this paper focuses on the 3D interpretation of the object. This is accomplished using a technique based on evidential reasoning and a wide range of rules and heuristics. The system is limited to the interpretation of objects composed of planar, spherical, and cylindrical surfaces. Experimental results are presented.

Converting paper-based engineering drawings into CAD model files is a tedious process, and automating the conversion of such drawings represents tremendous time and labor savings. We present a complete system which interprets such 2D paper-based engineering drawings, constructs solid models, and outputs 3D CAD-like files. While most existing systems assume that the engineering drawings have already been preprocessed to remove the dimensioning lines, this system is truly complete: It performs the detection of dimension sets, the extraction of object lines, and the assembly of 3D objects from object fines. A knowledge-based method is used to extract dimension sets and text from ANSI engineering drawings, and a graphics recognition procedure is used to extract complete object lines. In addition, a general-purpose dashed line recognition algorithm is presented. The focus of this paper, however is on the 3D interpretation of the object lines, which is accomplished using a technique that can accommodate auxiliary views in addition to the standard orthogonal set, along with planar spherical, and cylindrical object surfaces. Drawings with misaligned views and missing lines can also be interpreted. Extensive experimental results are presented

A method is presented for segmenting engineering drawings into views and identifying the corresponding view points. A set of 2.5D view-based coordinate systems is introduced as an intermediate between the 2D drawing-based system and the 3D object-based coordinates, and a formal technique is developed for constructing transformation matrices between coordinates. The method accommodates auxiliary views in addition to the standard orthogonal set, and the number of views and their positions need not be known a priori. Drawings with moderate errors in line placement and view alignment can also be handled. A rule based approach, using evidential reasoning, is applied for labeling the views.

A computer vision system to automatically analyze and assemble an image of the pieces of a jigsaw puzzle is presented. The system, called Automatic Puzzle Solver (APS), derives a new set of features based on the shape and color characteristics of the puzzle pieces. A combination of the shape dependent features and color cues is used to match the puzzle pieces. Matching is performed using a modified iterative labeling procedure in order to reconstruct the original picture represented by the jigsaw puzzle. Algorithms for obtaining shape description and matching are explained with experimental results