This paper presents an algorithm for extracting lines from hand drawings.It starts from contour pixel tracing, fits them into contour segments, and thenextracts skeleton lines from the contour segments. The algorithm finds all con-tours in one scan of the input matrix without detecting and marking multiplepixels. In line extraction, the method Elastic Contour Segment nacing is pro-posed which extracts lines by referring to the contour segments at both sides,overcoming noise and passing through blotted areas…

HChinese language deals with ideographic character input and output. TheX Window System (X11R5 or earlier versions) has the mechanism to outputHanzi (Chinese ideograph) through 2-byte encoding when the correspondingfolits are included, but it has no corresponding input feature. Because thereare as many as 16000 most in use Hanzis, more than one keystroke is needed toinput a single Hanzi by the traditional English keyboard. This paper proposesa Hanzi input architecture on the X Window System and a Chinese la…

A support system for form-correction of Chinese characters is developedbased upon a generation model SAM, and its feasibility is evaluated. SAMis excelIent as a model for generating Chinese characters, but it is difficult todetermine appropriate parameters because the use of calligraphic knowledge isneeded. By noticing that calligraphic knowledge of calligraphists is includedin their corrective actions, we adopt a strategy to acquire calligraphic knowl-edge by monitoring, recording and analyzing corrective …

A form evaluation system for brush-written Chinese characters is developed.Calligraphic knowledge used in the system is represented in the form of ruleswith the help of a data structure proposed in this paper. Reflecting the spe-cific hierarchical relations among radicals and strokes of Chinese characters,the proposed data structure is based upon a character model that can generatebrush-written Chinese characters on a computer. Through evaluation experi-ments using the developed system, it is shown that rep…

A new method to obtain spline outline description of Chinese font based onstroke extraction is presented. It has two primary advanages: (1) the qualityof Chinese output is greatly improved; (2) the memory requirement is reduced.The method for stroke extraction is discussed in detail and experimental resultsare presented.

Story understanding is one of the important branches of natural languageunderstanding research in AI techlliques. The story understanding approachbased on Story Parsing Grammar (SPG) involves that SPG is used to rep-resent different abstracting processes of stories with different levels in storyunderstanding and that the story understanding process is converted to therecognition process of stories using the syntactic parser of SPG. This kind ofstory understanding is cal1ed story parsing. In this paper, firs…

The equivalence- preserving transformation and normalization of types inobject-oriented databases are discussed. Specifically a normal form of typesbased on set-theoretic equivalence is proposed, rewrite rules which transformtypes into normal forms are presented, and the uniqueness of normal form andthe completeness of rewrite rules are proved. The emphasis of this work ison normal forms and corresponding rewrite rules. It provides a new formalapproach for the study of restructuring of database schema and o…

The development of the object-oriented paradigm has suffered from the lackof any generally accepted formal foundations for its semantic definition. Toaddress this issue, we propose the development of the logic-based semantics ofthe object-oriented paradigm. By combining the logic- with the object-orientedparadigm of computing first, this paper discusses formally the semantics of aquite purely object-oriented logic paradigm in terms of proof theory modeltheory and Aspoint theory from the viewpoint of logic. …

This paper investigates the semantics of conditional term rewriting systemswith negation (denoted by EI-CTRS), called constructor-based EI-model se-mantics. The introduction of "≠" in EI-CTRS make EI-CTRS more difficult tostudy. This is in part because of a failure of EI-CTRS to guarantee that thereexist least Herbrand models in classical logical point of views. The key idea ofEI-model is to explain that "t ≠ s" means that the two concepts representedby t and s respectively actually belong to distinguished …