Automatically Counting And Registering Hospital Patients Using Facial Detection And Recognition

Patients waiting for treatment is a common issue in health organizations.Even though improved services were created,still the issue remains,patients complain for long waits and queues.This thesis is based on a project provided and sponsored by a privately owned health organization that wants to reduce the issue of long waits at the patients’ waiting room of their organization.The company wants to advance its ability to handle this waiting issue via some new technological ideas or systems.These ideas or systems does not have to affect the waiting time directly,but can also provide services that from these services,they help reduce the waiting time of the patients.The Primary goal of the system is to reduce the patients’ waiting time before receiving his or her treatments.The secondary goal is to create some kind of services that will help speed up the staff’s work process and thereby reducing the patients’ waiting time.The project in this thesis will make use of two subfields in the Computer Vision arena.A system will be built composed of face detection and face recognition.The face detection component is used to detect number of faces in an image or video frame provided by the video camera in the patients’ waiting room.Number of faces detected equals number of possible patients waiting in the waiting room.The number of patients detected by the face detector will be sent to the staffs’ computers to keep them updated.This will help the staffs prepare the number of equipment and tools in advance according to the number of patients detected,thereby reducing waiting time.The project’s face recognition component will help in automatically registering new patients or unregistered patients by retrieving a new patient registration form and automatically filling in spaces that it acquired data on.In addition,it will be able to retrieve a registered patient’s profile and his or her medical history.From the tests that were carried out,the data proved that the project’s software module achieved its face detection goal by approximately 95%and face recognition by approximately 90%.