Color blindness is an anomaly which happened in retinal of eye(s) which prevent the patient to recognize or differentiate certain colors. The disability of the patient to recognize color is potential to cause problems to the patient in daily life. Color blind cannot be cured. Therefore, the only method to help color blind people to recognize or differentiate color is with a vision aid kit. In this work, color blind aid system for embedded platform based on Windows Embedded Standard 2009,.NET Framework, OpenCV library and EmguCV Wrapper developed. We create color recognition features implemented with augmented reality concept in the system. In this work, we developed a sound augmented reality concept and finger interaction between user and colored object. Test result according to system testing which has been done from 10 respondents resulting good feedback. The result of finger interaction test shows that the fingertip detection rate reaches 89.6% for skin classification method with HSV color space. Meanwhile, fingertip detection rate reaches 87.5% for skin classification method with YCbCr color space. Furthermore, color recognition rate achieved good result for the majority of tested color types.