欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
为了解决二维人脸识别准确度提升空间有限,三维人脸识别数据量大、识别速度慢的问题,提出了一种新的基于曲量场空间的人脸识别算法（Face Recognition based on Curved Space Field,FRCSF）.该算法首先检测彩色人脸图像内的面部凸凹信息,利用曲量子描绘凸凹域的渐变梯度特征,去除人脸彩色信息,降低三维信息量.然后以分散的曲量子群融合成曲量子空间.将曲量子空间进行边缘曲量子光滑衔接,组成曲量场空间.最后提取曲量场空间内的深度和维度信息,通过与曲量人脸库进行信息对比,判别出人脸身份.该算法抓住了人脸面部的凸凹特征,继而将凸凹特征采用具有空间连续性规律约束的曲量场进行描述,识别准确率较高,同时由于对三维人脸采用曲量子进行重建,数据量小,识别速度较快.大量实验表明,该算法既保存了二维人脸识别速度快的长处,又融入了三维人脸识别的局部三维信息,具有较高的识别性能.
英文摘要：
A novel approach to face recognition, which is based on the curved space field, is proposed to solve the problems of the limited accuracy promotion space of 2-D face recognition and the huge data quantity of 3-D face recognition, which causes the recognition speed to be slow. Firstly, the concavo-convex information in the color facial image is detected, the gradient feature of concavo-convex area is depicted with curved quantum, and the color information is replaced. Then all the sub-curved spaces, which are formed of curved quantum group, constitute the curved space field by linking up with each edge of curved quantum smoothly. Finally, recognition results are obtained by depth and dimension, both of which are extracted from curved space field. Experiments show that the proposed approach not only preserves the speed advantage of 2-D face recognition, but also integrates the part 3-D information of 3-D face recognition. The proposed approach achieves good recognition results.
同期刊论文项目
同项目期刊论文
期刊信息
位置：