Sharpness and Brightness Quality Assessment of Face Images for Recognition

Face image quality has an important effect on recognition performance. Recognition-oriented face image quality assessment is particularly necessary for the screening or application of face images with various qualities. In this work, sharpness and brightness were mainly assessed by a classification...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Ke Li, Hu Chen, Faxiu Huang, Shenggui Ling, Zhisheng You
Formato:	article
Lenguaje:	EN
Publicado:	Hindawi Limited 2021
Materias:	Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/ce6cde74129b41788ddbabcf1dbdbdfd
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:ce6cde74129b41788ddbabcf1dbdbdfd
record_format	dspace
spelling	oai:doaj.org-article:ce6cde74129b41788ddbabcf1dbdbdfd2021-11-08T02:36:49ZSharpness and Brightness Quality Assessment of Face Images for Recognition1875-919X10.1155/2021/4606828https://doaj.org/article/ce6cde74129b41788ddbabcf1dbdbdfd2021-01-01T00:00:00Zhttp://dx.doi.org/10.1155/2021/4606828https://doaj.org/toc/1875-919XFace image quality has an important effect on recognition performance. Recognition-oriented face image quality assessment is particularly necessary for the screening or application of face images with various qualities. In this work, sharpness and brightness were mainly assessed by a classification model. We selected very high-quality images of each subject and established nine kinds of quality labels that are related to recognition performance by utilizing a combination of face recognition algorithms, the human vision system, and a traditional brightness calculation method. Experiments were conducted on a custom dataset and the CMU multi-PIE face database for training and testing and on Labeled Faces in the Wild for cross-validation. The experimental results show that the proposed method can effectively reduce the false nonmatch rate by removing the low-quality face images identified by the classification model and vice versa. This method is even effective for face recognition algorithms that are not involved in label creation and whose training data are nonhomologous to the training set of our quality assessment model. The results show that the proposed method can distinguish images of different qualities with reasonable accuracy and is consistent with subjective human evaluation. The quality labels established in this paper are closely related to the recognition performance and exhibit good generalization to other recognition algorithms. Our method can be used to reject low-quality images to improve the recognition rate and screen high-quality images for subsequent processing.Ke LiHu ChenFaxiu HuangShenggui LingZhisheng YouHindawi LimitedarticleComputer softwareQA76.75-76.765ENScientific Programming, Vol 2021 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Computer software QA76.75-76.765
spellingShingle	Computer software QA76.75-76.765 Ke Li Hu Chen Faxiu Huang Shenggui Ling Zhisheng You Sharpness and Brightness Quality Assessment of Face Images for Recognition
description	Face image quality has an important effect on recognition performance. Recognition-oriented face image quality assessment is particularly necessary for the screening or application of face images with various qualities. In this work, sharpness and brightness were mainly assessed by a classification model. We selected very high-quality images of each subject and established nine kinds of quality labels that are related to recognition performance by utilizing a combination of face recognition algorithms, the human vision system, and a traditional brightness calculation method. Experiments were conducted on a custom dataset and the CMU multi-PIE face database for training and testing and on Labeled Faces in the Wild for cross-validation. The experimental results show that the proposed method can effectively reduce the false nonmatch rate by removing the low-quality face images identified by the classification model and vice versa. This method is even effective for face recognition algorithms that are not involved in label creation and whose training data are nonhomologous to the training set of our quality assessment model. The results show that the proposed method can distinguish images of different qualities with reasonable accuracy and is consistent with subjective human evaluation. The quality labels established in this paper are closely related to the recognition performance and exhibit good generalization to other recognition algorithms. Our method can be used to reject low-quality images to improve the recognition rate and screen high-quality images for subsequent processing.
format	article
author	Ke Li Hu Chen Faxiu Huang Shenggui Ling Zhisheng You
author_facet	Ke Li Hu Chen Faxiu Huang Shenggui Ling Zhisheng You
author_sort	Ke Li
title	Sharpness and Brightness Quality Assessment of Face Images for Recognition
title_short	Sharpness and Brightness Quality Assessment of Face Images for Recognition
title_full	Sharpness and Brightness Quality Assessment of Face Images for Recognition
title_fullStr	Sharpness and Brightness Quality Assessment of Face Images for Recognition
title_full_unstemmed	Sharpness and Brightness Quality Assessment of Face Images for Recognition
title_sort	sharpness and brightness quality assessment of face images for recognition
publisher	Hindawi Limited
publishDate	2021
url	https://doaj.org/article/ce6cde74129b41788ddbabcf1dbdbdfd
work_keys_str_mv	AT keli sharpnessandbrightnessqualityassessmentoffaceimagesforrecognition AT huchen sharpnessandbrightnessqualityassessmentoffaceimagesforrecognition AT faxiuhuang sharpnessandbrightnessqualityassessmentoffaceimagesforrecognition AT shengguiling sharpnessandbrightnessqualityassessmentoffaceimagesforrecognition AT zhishengyou sharpnessandbrightnessqualityassessmentoffaceimagesforrecognition
_version_	1718443138821914624

Sharpness and Brightness Quality Assessment of Face Images for Recognition

Ejemplares similares