CAJ | 학술논문

针对LPQ算子提取的纹理特征易受光照、噪声影响,提出了一种在Shearlet变换幅值域内提取局部相位量化标记直方图的人脸描述方法.首先采用 Shearlet 滤波器提取其对应不同方向、不同尺度的多个 Shearlet 幅值域图谱,然后按照平均值融合法将相同尺度不同方向的幅值域图谱融合到一起,并对融合图谱进行分块,分别采用LPQ算子标记幅值域图谱,最后由这些标记直方图形成的序列来描述人脸.在ORL、FERET和YALE人脸库中做了多组实验并分别取得了98%、95%及97.33%的识别率,充分验证了该方法的有效性.
침대LPQ산자제취적문리특정역수광조、조성영향,제출료일충재Shearlet변환폭치역내제취국부상위양화표기직방도적인검묘술방법.수선채용 Shearlet 려파기제취기대응불동방향、불동척도적다개 Shearlet 폭치역도보,연후안조평균치융합법장상동척도불동방향적폭치역도보융합도일기,병대융합도보진행분괴,분별채용LPQ산자표기폭치역도보,최후유저사표기직방도형성적서렬래묘술인검.재ORL、FERET화YALE인검고중주료다조실험병분별취득료98%、95%급97.33%적식별솔,충분험증료해방법적유효성.
@@@@Since the original Local Phase Quantization (LPQ) operator has a poor performance in extracting the texture features under illumination or noise effect, a method of face description is proposed which extracts the histogram sequence of Local Shearlet Phase Quantization (HLSPQ) from the magnitudes of Shearlet coefficients. First, In order to extract the multi-orientation information, the average fusion method is proposed to fuse the original Shearlet features of the same scale. Second, the fused image is divided into several units from which the local phase quantization operates is used to extract the local neighbor patterns. Then, the LPQ operates on each unit to extract the local neighbor patterns. Finally, the input face image is described by the histogram sequence extracted from all these region patterns. The experimental results on ORL, FERET and YALE face database can achieve high face recognition rate up to 98%,95%and 97.78%, which shows that the proposed method has better effect on improving the recognition rate.