 |
A Teacher from the Technology publishe a research about updating of basic components analysis to improve matching the characteristics of fixed-dimensional convert of recognize faces applications |
A Teacher from the Technology publishe a research about updating of basic components analysis to improve matching the characteristics of fixed-dimensional convert of recognize faces applications
The researcher said that the matching images based on learned characteristics, which is one of the most fundamental issues in computer vision tasks. As the growing some of those characteristics, the matching process sometimes become a bottleneck like a suffocating bottleneck.
The search displays a new method to improve the conversion of fixed characteristics scale. It has been tested the performance of proposed method on ORL AT & T database and it is found that the proposed APCA increased in it the characteristics learned when matching features of the faces in the pictures with the corresponding eigen faces and were compared with the result obtained from the original features of PCA. As a result, it led the use of PCA in the field waveform to reduce the size of the facial image input to the algorithm Sift, and thus led to increase the number of key points in the image of the face and allowed to obtain a better result when the comforting, in addition to the ease of implementation of the proposed method.
The research is published in the Journal of
International Journal of Artificial Intelligence and applications for smart devices Vol.4, NO.2 (2016), pp.9-18