Most university students are using the bus provided by the university's management to move from one place to another place. The analysis are required to improvise the quality of the of bus services such as the amount of passenger that using the bus and information of passengers such as gender. The objectives of this project are to develop face recognition system based on gender using Raspberry Pi 4 and Intel Neural Compute Stick 2 and to test and validate the performance of the developed system for face classification and passenger counting system. Also this system is able to store passenger information into Google Firebase Cloud with Internet of Things. This system is used Raspbian in Raspberry Pi 4 with the libraries that used for face classification and recognition such as OpenCV and OpenVINO. This project able to detect faces of the passengers soon as they ride the bus and determine gender of the passengers and count passengers according gender and the information of the passengers will stored in Google Firebase. There are some recommendation that need to be added in this project to improve efficiency of the system.

Security currently becomes a significant issue in public or private institutions in which various security system have been proposed and developed for some crucial processes such as person identifications, verification or recognition, especially for building access control. In this project, the design and development of a home security system for door accessing have been detailed, which uses face recognition and Near Field Communications (NFC). Thus, users can choose access to be used in the menu depends on face condition. In some face condition that face recognition unable to be identified, user can quickly decide to use NFC, which the user needed attach the NFC keychain to the NFC reader PN532. If NFC ID matches in the database, the solenoid was activated, and the door was open, meanwhile, if the user selects face recognition, the user required to take a picture on the camera in front of the door. Later, the recognition system identifies the face detection based on the database that has been created. The design displayed the output result of human face identity with a success rate of 48%. In conclusion, the objective, software and hardware of accessing door system is successfully developed and tested.