Abstract
This paper proposes a synthetic technique for human facial expression analysis based on locality preserving projections (LPP) and granular computing. The proposed method decreases the computational complexity of LPP and preserves the performance of LPP algorithm. In image processing, an image can be divided into various sizes of blocks, which are defined as image granules in this paper. The LPP algorithm is implemented to images with various image granules. By this strategy, the image dimension reduces quickly, which decreases the computational complexity. The experiments on three face databases are presented to show the performance of image granule LPP. Distribution of the facial pose and expression is shown and the computational complexities with different image granules are compared. Meanwhile, the order-preserving property of images is investigated by tracking the sequence of designated images. And the loss of image information is analyzed by image roughness, entropy and histogram. Furthermore, the parameter setting in LPP is discussed because of its non-ignorable affect on the experiment results. Finally, the method is applied to facial expression recognition.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: International Journal of Machine Learning and Cybernetics
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
