Face image recognition based on time series motif discovery

Publication Type:
Conference Proceeding
2012 IEEE International Conference on Granular Computing (GrC), 2012, pp. 72 - 77
Issue Date:
Full metadata record
Files in This Item:
Filename Description SizeFormat
2012004362OK.pdf832.3 kBAdobe PDF
This paper proposes a new face image recognition method based on time series motifs. Time series motifs are previously unknown, frequently occurring patterns in time series. In this paper, we make full use of time series motifs to discovery and recognize the face images. First, we convert face images into time series by extracting the time series features from three (primary color) vectors of the color face images. Second, we introduce our time series motif discovery algorithm to excavate the hidden motifs from the time series of color face images. Third, we find the relationship among these generated time series. At last, experiments are conducted to verify the feasibility and effectiveness of our algorithm, and we can safely draw a conclusion that our approach in the face recognition field is valid and successfully realizes the face image recognition from a new perspective.
Please use this identifier to cite or link to this item: