Character Font Research Articles

Proof-reading on VDUs

Abstract Two experiments are reported which compared proof-reading performance across three different modes of presentation. The results of Experiment 1 indicated that proof-reading accuracy was significantly worse on a VDU than on paper, with a photograph of the VDU display producing intermediate performance. It was also demonstrated that substitutions of visually similar errors were harder to detect than other error types. The results of Experiment 2 replicated these findings, but failed to find any difference between twin-column and single-column presentation. The findings are interpreted as indicating that character font may be a major factor in the poor performance with the VDU. It is suggested that the present experimental paradigm could be a useful assessment tool for the evaluation of display design.

STATISTICALLY EQUIVALENT BLOCK (SEB) CLASSIFIER: ITS ASYMPTOTIC PROPERTY AND PROSPECTS

A mathematical model of statistically equivalent block (SEB) classifier has been discussed, the decision regions (blocks) of which being as general as Borel measurable sets in Rn. It has been proved that the condition for the recognition rate of SEB classifier to converge to Bvyes‘ rate in probability is that log n/kn→0, n being the training sample size and kn the number of sample patterns contained in each block. This result is superior to those yet claimed in other papers. In contrast to traditional “two-step” method of classifier analysis, a one-step scheme without density estimation of each class has been developed in this work to strengthen the mathematical proof. A decision tree implementation of SEB classifier with corresponding algorithms was proposed. A simulation experiment on SEB classifier was conducted, which 1) justifies the above theoretical result; 2) suggests small value of kn be preferred in SEB classifier training, which coincides with the theorems given in this paper; 3) shows prospects of SEB classifier in pattern recognition applications. As a nonparametric classifier, SEB classifier is better than the K–NN rules with respect to time and space complexities. In fact, it has been applied to character font recognition in another work of the author.

Why is Reading Slower from CRT Displays than from Paper

Experiments, including our own (Gould et al., 1984; 1986), have shown that people read more slowly from CRT displays than from paper. A series of experiments shows that the explanation centers on the image quality of the characters. Reading speeds, equivalent to those on paper, have been found for CRT displays containing character fonts that resemble those on paper (rather than dot matrix fonts, for example), that have a polarity of dark characters on a light background, that are anti-aliased (i.e., contain grey level), and that are shown on displays with relatively high resolution (e.g., 1000 × 800).

A computer aided design system for artistic Chinese fonts

Abstract The design of artistic fonts for characters is used for producing special visual effects via the shape and color changes on the character fonts. The various changes on the shapes of fonts can be achieved by changing the boundary curves of fonts which are represented by spline functions. In this paper, we focus our attention on the Chinese character fonts. A mathematical simulation of simplifying two dimensional and three dimensional surface projection is provided. The implementation of the system is presented with illustrative examples.

Why reading was slower from CRT displays than from paper

Experiments, including our own (Gould et al., 1982; 1984; 1986), have shown that people read more slowly from CRT displays than from paper. Here we summarize results from a few of our fifteen experiments that have led us to conclude that the explanation centers on the image quality of the CRT characters. Reading speeds equivalent to those on paper were found when the CRT displays contained character fonts that resembled those on paper (rather than dot matrix fonts, for example), had a polarity of dark characters on a light background, were anti-aliased (e.g., contained grey level), and were shown on displays with relatively high resolution (e.g., 1000 x 800). Each of these variables probably contributes something to the improvement, but the trade-offs have not been determined. Other general explanations for the reading speed difference that can be excluded include some inherent defect in CRT technology itself or personal variables such as age, experience, or familiarity at reading from CRT displays.

Why reading was slower from CRT displays than from paper

Experiments, including our own (Gould et al., 1982; 1984; 1986), have shown that people read more slowly from CRT displays than from paper. Here we summarize results from a few of our fifteen experiments that have led us to conclude that the explanation centers on the image quality of the CRT characters. Reading speeds equivalent to those on paper were found when the CRT displays contained character fonts that resembled those on paper (rather than dot matrix fonts, for example), had a polarity of dark characters on a light background, were anti-aliased (e.g., contained grey level), and were shown on displays with relatively high resolution (e.g., 1000 x 800). Each of these variables probably contributes something to the improvement, but the trade-offs have not been determined. Other general explanations for the reading speed difference that can be excluded include some inherent defect in CRT technology itself or personal variables such as age, experience, or familiarity at reading from CRT displays.

TEXT: Publication-quality characters come to APL graphics

An APL workspace named TEXT has been developed to generate publication-quality characters for use with several APL plotting packages, both on the mainframe and on the PC. TEXT is written in APL and is very simple to use. The fonts supplied are based on the Hershey fonts - a well known public-domain digitization of over 20 character fonts. The user is provided with several functions that construct the vectors for plotting a character string (vectors signifying move and draw locations) — including lower-level functions that are more flexible for certain applications. Some of these functions are: rotating, scaling, justifying, underlining and changing style of text strings. Greek, Italic, Russian, Gothic, Math, Weather, and Music symbols are just a few of the character sets available. A modified version is described for APL2 1 , which uses mixed arrays as input.

Chinese character processing systems in China

A general review of Chinese character processing systems developed in China is presented. Three main categories of existing entry schemes, i.e. character entry, encoding and component keying-in, are described. After examing the diversified encoding schemes based on phonetics, shape and phonetics-shape, it is pointed out that computer-aided component keying-in scheme is more ergonomical and promising in the near future. Section of Main Features gives an overall picture of the Chinese editor, character font, software for bilingual processing, display and printer for Chinese characters. Then the standardization of the character font, code for information exchange, machine code as well as Chinese character terminal are discussed. Finally, an overview of the applications of Chinese character processing systems in the enterprises, offices, facilities for public service and Chinese medical diagnosis are given.

Microprocessor interfaces to NCR bytewide E 2PROMS

NCR 52832 (Figure 1) and 52864 5-V-only E 2PROMS can be used in applications where occasional insystem programmability and long-term data retention without power (non-volatility) are needed. Applications include the storage of programmable character fonts in intelligent terminals, storage of data or firmware in remote systems that need periodic updating, user-programmable video games, and RAM backup.

A character set for molecular structure display

A character set is presented with which unscaled 2D molecular structural diagrams can be constructed. A minimal subset has been implemented in ROM on CBM microcomputers, whilst the complete set has been defined as a character font on an ICL PERQ computer. Structural diagrams have also been output on dot-matrix and daisy-wheel printers using appropriate software to print the special characters. The character set provides a simple means of displaying molecular structures on inexpensive microcomputers, on printers and on hardware offering user-definable character fonts.

Character Display on a CRT

Character fonts on raster scanned display devices (e.g., the home television receiver) are usually represented by arrays of bits and displayed as a matrix of black and white (or color) dots. In VIDEOTEX systems, due to the limitations on bandwidth, number of scanning lines, and the interlace of the broadcast television standard, each frame is allowed to contain about 21 rows of 40 characters each; each character occupying a rectangular area of 12 picture elements by 20 lines. The character sets that have traditionally been used in VIDEOTEX systems, have not used the multilevel display capabilities of the television receiver. Due to interlace, this results in flicker at every sharp color vertical transition. In this note, we show how the multilevel capabilities of a television receiver can be used by nonlinearly filtering the characters generated by a conventional VIDEOTEX character generator such that each vertical transition is made less sharp. Although the characters are more blurred due to filtering, the apparent resolution of the display is increased allowing smaller fonts to be used. Flicker is reduced to a large extent, thus, decreasing the fatigue due to prolonged viewing of the display. A font occupying 9X16 rectangular area per character is proposed. It results in about 65 percent more characters per 'page-screen' compared to the present VIDEOTEX system. The hardware required for such nonlinear filtering is minor, since, storage of multilevel intensities of the characters is not required due to local processing at the receiver.

A system model for Chinese character fonts generation

Abstract A system model for the generation of high‐quality Chinese characters is presented. A software tool called the ACCFONT system has been implemented to solve a particular inherent problem in designing the Chinese character fonts, that is, the variation in the shape of a stroke varies from character to character. In the model, the designer constructs Chinese character fonts by a numerical‐data‐oriented command language, and the ACCFONT system produces nonnumerical decisions to modify the shape of strokes properly. In this paper, we show how the ACCFONT system is modeled. Also, some examples are given to illustrate its important features.

The display of characters using gray level sample arrays

Character fonts on raster scanned display devices are usually represented by arrays of bits that are displayed as a matrix of black and white dots. This paper reviews a filtering and sampling method as applied to characters for building multiple bit per pixel arrays. These arrays can be used as alternative character representations for use on devices with gray scale capability. Discussed in this paper are both the filtering algorithms that are used to generate gray scale fonts and some consequences of using gray levels for the representation of fonts including: 1. The apparent resolution of the display is increased when using gray scale fonts allowing smaller fonts to be used with higher apparent positional accuracy and readability. This is especially important when using low resolution displays. 2. Fonts of any size and orientation can be generated automatically from suitable high precision representations. This automatic generation removes the tedious process of “bit tuning” fonts for a given display.

Data compression of Chinese character patterns

Data compression methods of Chinese character patterns are presented, which are mainly used for the output printing from computer. Various methods have been tried aiming at the smallest amount of data storage, the reproduction of artistic character fonts, and so on. Among others, dot pattern representation is the most popular methods actually used for representing Chinese characters. There are many other methods, such as contour coding, skeleton representation, and syntactic construction from subpatterns. The details of these are explained with a brief introduction of some general properties of Chinese characters per se used in Japan.

Remarks about printing and displaying some non-latin characters

Most typewriters, electronic printers and display systems were designed initially, specifically for Latin characters. This has delayed the introduction of conversational and business terminals in regions where the written language calls for using characters of an essentially different nature. A typical case of this situation occurs for the Arabic group of languages upon which the paper will place special emphasis.It is possible to adapt typewriters and printers into devices, useful for the given purpose, but such adaptations are always "ad hoc", that is for a particular given language, and often involve using trade-offs that are sometimes inconvenient or aesthetically unacceptable. For most electronic display systems such an adaptation even proves to be impossible in practice. The problems that arise are due to the fundamental structural differences between the typing of Latin and non-Lation texts. These differences relate mainly to (a) the necessity of joining some or all of the letters in a word, (b) the impossibility of incorporating all characters in a module of a fixed length, (c) the variations a character undergoes, depending on its place in a word and (d) the need for composing, in some cases, the characters out of different elementary signs.Most of these difficulties can and have been overcome to some degree by generating the characters by a dot-matrix. The matrix of 5 × 7 dots commonly used for Latin script is insufficient for the representation of Arabic characters. An example is presented where using matrices of 7 × 7 and 14 × 7 dots leads to quite satisfactory representations. Patterns of 14 × 7 dots can of course be treated as a double impression with a 7 × 7 matrix. Other solutions have been used, that call for the typist to compose some characters out of several, partial impressions.A simple, elegant and convenient solution is proposed for printing as well as for displaying. It is achieved by using a single column of dots as the basic representational unit. Since such a configuration does not imply any modularity or periodicity foreign to the language to be represented, it can be easily sequenced so as to achieve any required sequence of printed characters, as long as the vertical resolution of 7 dots is sufficient. Additional resolution can of course be provided.It is concluded that on the basis of the proposed procedure printers and display units can be developed, that would be controlled by a microprocessor and could be used with identical hardware for representing many different scripts or for different styles of typing belonging to the same language. The specificity of the display unit or the printer with respect to its character font would reside in a set of data (character file) and a driver program stored in a ROM-memory. The character file could be identical for a display unit as for the printer.Additional advantages of the proposed procedure are: (a) The presence of a microprocessor in the system and the"intelligence" it provides can be used to alleviate some of the tasks of the typist and to achieve a better performance as a terminal. (b) The printed text would be well adapted for treatment with automatic text-readers.

An automatic measurement device for the evaluation of the print quality of printed characters

When using OCR equipment (OCR = Optical Character Recognition) it is necessary to define shape and print quality of the characters in national and international standards. The paper gives a review of the standardization of the character fonts OCR A and OCR B for optical character recognition. All relevant parameters are discussed which define the print quality of printed characters. A critical review of two measurement methods used so far for the evaluation of print quality parameters is given. The improvement of the measurement methods led to the development of an automatic measurement device which is described in detail. The automatic measurement device consists of a high precision scanning device to digitize the printed characters and a computer program for the evaluation of the data. The mechanical construction and the properties of the scanning device are explained. The computer program consists of two main parts for preprocessing of the scanned data and for the evaluation of the different print parameters. Test results are reported and possible applications of the measurement device are discussed.

A pattern identification system using linear decision functions

This paper is concerned with application of linear decision functions to the pattern identification problem and describes an experimental pattern recognition system for the magnetic ink character font now used in the banking industry. The system is based on a linear decision function determined by means of a variant of an adaptive training technique due to Rosenblatt. The system has been partially implemented (in part, through simulation with aid of a digital computer and, in part, by hardware) and experimental results in using the system are reported.