Este projeto utiliza a linguagem de programação C++-17 em conjunto com a biblioteca OpenCV para realizar a classificação de faces utilizando a técnica de PCA.
10 componentes principais, acurácia: 92%.
15 componentes principais, acurácia: 94%.
20 componentes principais, acurácia: 95%.
No Linux instale as seguintes bibliotecas:
sudo apt install make g++ binutils libopencv-dev
Para compilar:
bash build.sh
Para executar:
./PCA