CAJ | 학술논문

提出一种联合局部三值模式（LTP）和局部相位量化（LPQ）的人脸识别方法，该方法首先对预处理后的人脸分别采用LTP算子、LPQ算子提取各自的直方图，然后，通过特征选择将两者联合为LTP＿LPQ直方图，最后通过最近邻分类器进行分类识别。在ORL和YALE标准人脸数据库上的实验表明，该方法能有效地提高人脸的识别率。
제출일충연합국부삼치모식（LTP）화국부상위양화（LPQ）적인검식별방법，해방법수선대예처리후적인검분별채용LTP산자、LPQ산자제취각자적직방도，연후，통과특정선택장량자연합위LTP＿LPQ직방도，최후통과최근린분류기진행분류식별。재ORL화YALE표준인검수거고상적실험표명，해방법능유효지제고인검적식별솔。
A method of face recognition based on local ternary pattern(LTP) and local phase quantization(LPQ) was proposed .Firstly ,each histogram by LTP operator and LPQ operator was extracted respectively based on the pretreated faces ,then both histograms were fused into the LBP_LPQ histogram by feature selection ,finally they were identified by the nearest neighbor classifier .Experiments based on the ORL and YALE standard face database illustrated that this method can effectively improve the recognition rate of face .