Occluded Face Recognition Based on the Deep Learning
Authors:
- 1 June 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 793-797
Abstract
With the aggravation of social public security problems, the face recognition without occluded parts can no longer meet the needs of modern society. All kinds of face recognitions in complex environment need to be realized in the real situations. The paper proposed a new method to recognize the occluded face based on the deep learning. The face recognition model is trained and learned based on convolution neural network of the deep learning, which has strong robustness to illumination difference, facial expression change and facial occlusion. Through a large number of experimental tests and result analysis, the occluded face recognition rate can reach up to 98.6%. Therefore, this method proposed in the paper realizes face recognition with occlusion in complex environment and meets the needs of practical applications.Keywords
This publication has 9 references indexed in Scilit:
- 3D modelling of small object based on the projector‐camera systemKybernetes, 2012
- Job-Level Proof Number SearchIEEE Transactions on Computational Intelligence and AI in Games, 2012
- Bitboard knowledge base system and elegant search architectures for Connect6Knowledge-Based Systems, 2012
- TAAI 2011 Computer-Game TournamentsICGA Journal: The Journal of the Computer Games Community, 2011
- Two-Stage Monte Carlo Tree Search for Connect6IEEE Transactions on Computational Intelligence and AI in Games, 2011
- Job-Level Proof-Number Search for Connect62011
- Relevance-Zone-Oriented Proof Search for Connect6IEEE Transactions on Computational Intelligence and AI in Games, 2010
- Construction of opening book in connect6 with its application2009
- Enhancements of proof number search in connect62009