电子计算机汉字信息处理中的编码问题和自动识别探讨

INVESTIGATION OF THE CODING PROBLEM ABOUT THE ELECTRONIC COMPUTER CHINESE CHARACTER INFORMATION PROCESSING AND AUTOMATIC RECOGNITION

目前,解决汉字输入的方法,基本上可以分为两大类。一类是将汉字用人工进行编码,再将编码输入计算机内。另一类是用计算机自动识别汉字,即通过专用设备将汉字输入计算机,然后对输入的汉字自动进行分析、识别,自动进行编码。本文对国内提出的一些编码方法作了一些评价,并提出一种拚音字、词代码,作为汉字人工输入计算机的手段。这种代码具有不受汉字字形变化影响,以及良好的可读性能。当考虑计算机处理文字信息的速度与人工键盘输入汉字的速度不相匹配时,本文还以笔划为基本识别特征,对汉字的自动识别方法进行了一些探讨。

At present, the method for solving the Chinese character input problem may be basically divided into two types.One type is manual code and this code is inputted into the computer. And the other type is automatic recognition with the computer, that is to say,the Chinese character is inputted by dedicated device, and then the computer carrys out the automatic analysis, recognition and coding.This article makes some evaluation for the method of some domestic coding, and provide a code of phonetic character and words, this code can be treated as a means of manual input computer. This code is independent of Chinese character's shape, and has a fine readable performance.Besides, the speed for computer to treat literal information and for manual keyboard to input the Chinese character is not matching, therefore, for the automatic recognition of Chinese character some investigations by means of the strokes of a Chinese character as basic characteristic of recognition were made in this article.