Improving accuracy of facial recognition systems for distant targets

摘要

This work is aimed at improving the accuracy of the PCA-based facial recognition system for distant targets. Since the faces captured from distant targets are inevitably with lower resolution in contrast to the enrolled faces in the database, enhancing the effective resolution will be accordingly adopted to improve the accuracy. This work will category the approaches and experimentally compare the performance of them, which provides a valuable recommendation for the design of a practical facial recognition system.