当前位置: X-MOL 学术Multimed. Tools Appl. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Proposed framework for cancelable face recognition system
Multimedia Tools and Applications ( IF 3.6 ) Pub Date : 2021-01-17 , DOI: 10.1007/s11042-020-10291-z
H. I. Ashiba

This paper suggests two novel presented cancellable biometric realization approaches recognition and template protection. In the suggested scheme, the A Trous Transform (AT) algorithm is applied on the face images. Then the AT divides the image into seven subbands. The resultant map is encrypted with the Homomorphic Filtering Masking (HFM) encoding algorithm is utilized for cancelable face recognition system. Then the second HFM utilized is produced from the image. This technique can be used to advance a frequency domain procedure for making this system for biometric template protection. The second algorithm presents a new technique to detect the features for Facial Expression Recognition (FER). This technique is established on segmentation process by Canny Edge Detection (CD) and Hough Transform (HT) with the number of feature points as the key parameters for classification. This algorithm is set up analysis the FER with the number of HT feature points as the key parameters for classification. Simulation results using evaluation metrics False Positive Rate (FPR), False Negative Rate (FNR), Equal Error Rate (EER), Receiver Operating Characteristic (ROC) and Area under ROC (AROC) prove that the first proposed cancelable biometric technique with the second key are best with comparing the other keys.The obtained results clear that the second suggested technique has sucesseded in detection the features of FER for sad, happy, neutral, angry, disgust, fear, surprise cases and the face positions.



中文翻译:

提议的可取消人脸识别系统框架

本文提出了两种新颖的可取消生物识别实现方法:识别和模板保护。在建议的方案中,将Trous变换(AT)算法应用于面部图像。然后,AT将图像划分为七个子带。生成的地图使用同态滤波蒙版(HFM)编码算法加密,用于可取消的面部识别系统。然后,从图像中生成第二个HFM。此技术可用于推进频域过程,以使该系统用于生物特征模板保护。第二种算法提出了一种检测面部表情识别(FER)特征的新技术。该技术是基于Canny Edge Detection(CD)和Hough Transform(HT)在分割过程中建立的,以特征点的数量为分类的关键参数。建立了以HT特征点数量为分类关键参数的FER分析算法。使用评估指标误报率(FPR),误报率(FNR),等误码率(EER),接收器工作特征(ROC)和ROC下面积(AROC)进行的仿真结果证明,第一个提出可取消的生物特征技术,第二个提出可取消的生物特征技术获得的结果清楚地表明,第二种建议的技术已成功地用于检测FER的悲伤,快乐,中立,愤怒,厌恶,恐惧,意外情况和面部位置的特征。建立了以HT特征点数量为分类关键参数的FER分析算法。使用评估指标误报率(FPR),误报率(FNR),等误码率(EER),接收器工作特征(ROC)和ROC下面积(AROC)进行的仿真结果证明,第一个提出可取消的生物特征技术,第二个提出可取消的生物特征技术获得的结果清楚地表明,第二种建议的技术已成功地用于检测FER的悲伤,快乐,中立,愤怒,厌恶,恐惧,意外情况和面部位置的特征。建立了以HT特征点数量为分类关键参数的FER分析算法。使用评估指标误报率(FPR),误报率(FNR),等误码率(EER),接收器工作特征(ROC)和ROC下面积(AROC)进行的仿真结果证明,第一个提出可取消的生物特征技术,第二个提出可取消的生物特征技术获得的结果清楚地表明,第二种建议的技术已成功地用于检测FER的悲伤,快乐,中立,愤怒,厌恶,恐惧,意外情况和面部位置的特征。

更新日期:2021-01-18
down
wechat
bug