基于指数爆炸性和相似度的三角形识别方法
首发时间:2023-01-09
摘要:为了解决在模糊模式识别下三角形识别结果不准确的问题,本文基于指数爆炸性和三角形相似度,提出两种新的三角形识别方法.第一种改进思路是将对数函数放缩至可控的范围,使其成为分段函数的形式,在一定程度上保留了数据.第二种改进思路利用正弦定理、均值不等式、琴生不等式和三角形相似、平移和旋转等知识,使三角形固定于一单位圆中,从面积、边长和角度等方面寻找任意一个三角形与具有一定特征的三角形的相似度,最终重新定义了五种类型的隶属函数.测试数据结果表明,新方法识别三角形准确率高,弥补了原有识别方法的不足,具有很好的应用前景.
For information in English, please click here
A triangle identification method based on exponential explosivity and similarity
Abstract:n order to solve the problem of inaccurate triangle recognition results in fuzzy pattern recognition, this paper proposes two new triangle recognition methods based on exponential explosion and triangle similarity. The first idea is to reduce the logarithm function to a manageable range, making it a piecewise function, which preserves the data to a certain extent. The second improvement ideas by using sine theorem, mean inequality, the born inequality and triangle similarity, translation and rotation, such as knowledge, make the triangle fixed in a unit circle, from the aspects such as area, length and Angle for an arbitrary triangle and has certain characteristics of the similarity of the triangle, finally has redefined the five types of membership function. Experimental results show that the new method has high recognition accuracy, makes up for the shortcomings of the original recognition methods, and has a good application prospect.?????
Keywords: Triangles Fuzzy pattern recognition Retraction Law of sines Similarity
基金:
引用
No.****
同行评议
勘误表
基于指数爆炸性和相似度的三角形识别方法
评论
全部评论0/1000