一种基于核典型相关分析的煤炭矸石鉴别方法

摘要

提出了一种利用核典型相关分析(KCCA)来抽取煤炭矸石的非线性鉴别特征,并用其进行煤炭矸石自动识别的方法.实验表明,对小样本煤炭矸石图像,提出的方法可以得到较好的识别性能,再结合机械自动化技术即可以达到煤矸自动分选的目的.%This paper presents a method based on the KCCA(Kernal Canonical Correlation Analysis) to extract the nonlinear characteristics of the waste rock,and apply it to recognise the waste rock automatically. The experiments show that this method can get good recognition performance with small sample waste rock images. When combined with mechanical automation technology, this method can achieve the purpose of the waste rock sorting.