About this document

Enhancing CNN Face Recognition Accuracy by grazzymaya02 is a document available to read on EtoBox.

The document discusses preprocessing techniques that can improve the accuracy of CNN-based face recognition systems. Experiments on the Extended YALE B and FERET databases showed that applying preprocessing methods like SQI, HE, LTISN, GIC and DoG to CNN models improved accuracy. On the Extended YALE B database, accuracy improved from 96.2% without preprocessing to 99.8% with preprocessing. On the FERET database, accuracy improved from 71.4% without preprocessing to 76.3% with preprocessing. The goal of pre

Author

grazzymaya02

Language

EN