在模式识别中由于刻画事物的特征值具有不确定性,而区间数又能够有效地描述这种不确定性。因此,为了更好地研究这类模式识别问题,本文将区间数用于模式识别中,基于区间数的中点和宽度引入区间数的距离,依据择近原则,确定待识别对象的归属,最后通过算例验证其有效性。
There is often an uncertainty of eigenvalue description in pattern recognition which can be effectively described by interval numbers. Therefore, in order to better study this kind of pattern recognition problems, this paper applies interval numbers to pattern recognition. Based on the midpoint and the width of the interval numbers, it introduces the interval numbers distance, confirms the ad- scription of the object to be identified under the principle of selecting the nearest, and verifies its effectiveness by an example at last.