An Electronic Business Card System with Uniform Local Binary Patterns Algorithm for Real-Time Individual Face Recognition

An electronic business card system based on mobile devices with the function of real-time face recognition is proposed in this article. With mobile devices, users can record and exchange information on their friends, including images. Users only need to take a picture of a person with the camera of the mobile device, and the system finds the personal information timely. A face recognition algorithm based on Local Binary Patterns Histograms (LBPH) is implemented in the application program (APP) which can be deployed on the mobile phone. For fast runtime performance, the image is down-sampled and divided into blocks, and the LBPH of individual blocks are concatenated to construct the final feature vector of the image. Finally, the feature vector is used to find the best-matching image in the dataset. The function of face recognition is completed in 1.5 s on average, the maximum tilted angle of a face image detected is 42°, and the longest distance recognized is up to 11 ms. Moreover, the average success rate of face recognition reaches over 90% under a normal intensity of light, which is sufficient to cope with the social needs of the general situation.