Abstract

Aiming at the application scene of face image recognition in audit investigation, firstly, a face data set with corresponding environmental interference is simulated to increase the diversity of data; in the preprocessing stage, the improved method of adaptive histogram illumination balance and simulated glasses covering is used for image enhancement; in the model training stage, an optimal weight re overloading model training algorithm is proposed. The experimental results show that the accuracy, robustness and efficiency of face recognition in the application scene are improved by the improvement of image enhancement preprocessing and the model training of optimal weight reloading.

Details

Title
Research and Application of Facial recognition Algorithm in Audit Investigation
Author
Wang, Hairong 1 ; Li, Weibo 1 ; Wan, Quan 1 ; Yan, Hua 1 ; Xiang, Rui 1 

 School of Computer Science & Engineering, Wuhan Institute of Technology, Wuhan, China; Hubei Key Laboratory of Intelligent Robot, Wuhan Institute of Technology, Wuhan, China 
Publication year
2021
Publication date
Apr 2021
Publisher
IOP Publishing
ISSN
17426588
e-ISSN
17426596
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2511968645
Copyright
© 2021. This work is published under http://creativecommons.org/licenses/by/3.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.