Line drawing interpretation is a challenging area with enormous practical potential. At present, many companies throughout the world invest large amounts of money and human resource in the input of paper drawings into computers. The technology needed to produce an image of a drawing is widely available, but the transformation of these images into more useful forms is an active field of research and development. Machine Interpretation of Line Drawing Images - describes the theory and practice underlying the computer interpretation of line drawing images and - shows how line drawing interpretation systems can be developed. The authors show how many of the problems can be tackled and provide a thorough overview of the processes underpinning the interpretation of images of line drawings.

This book solves a long-standing problem in computer vision, the interpretation of line drawings and, in doing so answers many of the concerns raised by this problem, particularly with regard to errors in the placement of lines and vertices in the images. Sugihara presents a computational mechanism that functionally mimics human perception in being able to generate three-dimensional descriptions of objects from two-dimensional line drawings. The objects considered are polyhedrons or solid objects bounded by planar faces, and the line drawings are single-view pictures of these objects. Sugihara's mechanism has several potential applications. It can facilitate man-machine communication by extracting object structures automatically from pictures drawn by a designer, which can be particularly useful in the computer-aided design of geometric objects, such as mechanical parts and buildings. It can also be used in the intermediate stage of computer vision systems used to obtain and analyze images in the outside world. The computational mechanism itself is not accompanied by a large database but is composed of several simple procedures based on linear algebra and combinatorial theory. Contents:Introduction. Candidates for Spatial Interpretation. Discrimination between Correct and Incorrect Pictures. Correctness of HiddenPart-Drawn Pictures. Algebraic Structures of Line Drawings. Combinatorial Structures of Line Drawings. Overcoming Superstrictness. Algorithmic Aspects of Generic Reconstructibility. Specification of Unique Shapes. Recovery of Shape from Surface Information. Polyhedrons and Rigidity. Kokichi Sugihara is Professor in the Department of Mathematical Engineering and instrumentation Physics, Faculty of Engineering, the University of Tokyo, Tokyo, Japan. Machine interpretation of Line Drawingsis included in The MIT Press Series in Artificial Intelligence, edited by Patrick Henry Winston and Michael Brady.

A study of multispectral image processing and pattern recognition. It covers: geometric and orthogonal moments; minimum description length method for facet matching; an integrated vision system for ALV navigation; fuzzy Bayesian networks; and more.

The image analysis community has put much effort into developing systems for the automatic reading of various types of documents containing text, graphic information, and pictures. A closely related but much more problematic task is the reading and interpretation of line drawings such as maps, engineering drawings, and diagrams. This book considers the problem in detail, analyzes its theoretical foundations, and analyzes existing approaches and systems.

Based on the author’s considerable research, this book contains state-of-the-art reviews of work in drawing interpretation and discrete optimization. It covers both drawings of polyhedral objects as well as complex curved objects.

1 Thisbookcontainsrefereedandimprovedpaperspresentedatthe5thIAPR - ternational Workshop on Graphics Recognition (GREC 2003). GREC 2003 was held in the Computer Vision Center, in Barcelona (Spain) during July 30–31, 2003. TheGRECworkshopisthemainactivityoftheIAPR-TC10,theTechnical 2 Committee on Graphics Recognition . Edited volumes from the previous wo- shops in the series are available as Lecture Notes in Computer Science: LNCS Volume 1072 (GREC 1995 at Penn State University, USA), LNCS Volume 1389 (GREC 1997 in Nancy, France), LNCS Volume 1941 (GREC 1999 in Jaipur, India), and LNCS Volume 2390 (GREC 2001 in Kingston, Canada). Graphics recognition is a particular ?eld in the domain of document ana- sis that combines pattern recognition and image processing techniques for the analysis of any kind of graphical information in documents, either from paper or electronic formats. Topics of interest for the graphics recognition community are: vectorization; symbol recognition; analysis of graphic documents with - agrammatic notation like electrical diagrams, architectural plans, engineering drawings, musical scores, maps, etc. ; graphics-based information retrieval; p- formance evaluation in graphics recognition; and systems for graphics recog- tion. Inadditiontotheclassicobjectives,inrecentyearsgraphicsrecognitionhas faced up to new and promising perspectives, some of them in conjunction with other, a?ne scienti?c communities. Examples of that are sketchy interfaces and on-line graphics recognition in the framework of human computer interaction, or query by graphic content for retrieval and browsing in large-format graphic d- uments, digital libraries and Web applications. Thus, the combination of classic challenges with new research interests gives the graphics recognition ?eld an active scienti?c community, with a promising future.