摘要
针对传统的基于主分量分析的模式识别很难获得很高的识别率的问题,提出了一种基于主分量分析的融合识别方法。采用D- S证据理论对基于两种K- L 变换的主分量分析法提取的低维特征进行融合识别。交通标志的形状识别实验表明了该融合识别算法降低特征维数的同时有效地提高了识别率。
In allusion to the question that it can not get both recognition rates and character dimensions in the pattern recognition at the same time, it proposes a method of fusion recognition based on PCA. It uses Dempster-Shafer theory to make fusion recognition to two extracted low dimension characters, which are based on two different K-L transforms. The shape recognition test of traffic signs has indicated this method of fusion recognition can reduce character dimensions and advance the recognition rate effectively.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2004年第z3期440-442,共3页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金资助项目(60175011
60375011)
安徽省重点科研项目(03021012)
安徽省优秀青年科技基金资助项目(03042044)。
