Biometrics is computer technology for the recognition of oneself through certain body parts. The use of pupil eyes as a feature of a person is the result of extracting facial features using geometry features, measurement of distance between eyes is still manually so that it requires a system that directly displays the results of measurements to obtain PD values that can be directly used to identify someone. This study designed a face recognition system through the MATLAB GUI to simplify the process of image capture and processing. The image is taken using a smartphone that is connected directly to a portable computer wirelessly through the Internet Protocol Camera system. The number of image samples used amounted to 50 samples for the database and 100 samples for the test data consisting of 10 people with the composition of 5 men and 5 women. Image processing uses the Viola-Jones method to detect faces in the second area of the eye and uses the Euclidean distance method to find the pupillary distance value which is then followed by image grouping based on that range of values and ends with facial recognition system testing. The test uses 100 samples of image data consisting of 5 men and 5 women with each of the 10 image data. Based on the results of these tests, the accuracy of face recognition system is obtained by 73% and the accuracy value based on gender for men is 78% and women are 68%. With face matching speed of about 2-3 seconds..