Development of drone-based solution for medicine delivery using face recognition and guidance landing system / Mohamed Osman Baloola Mohamed
This thesis presents the Development OF Drone-Based Solution for Medicine Delivery Using Face Recognition and Guidance Landing System. The drone is equipped with a Sanitiser Unit (SU) and a secure Delivery Box (DB) for contactless medication delivery using face recognition and the GLS. The dev...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2022
|
| Subjects: | |
| Online Access: | http://studentsrepo.um.edu.my/14862/1/Mohamed_Osman.jpg http://studentsrepo.um.edu.my/14862/8/osman.pdf http://studentsrepo.um.edu.my/14862/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This thesis presents the Development OF Drone-Based Solution for Medicine Delivery
Using Face Recognition and Guidance Landing System. The drone is equipped with a
Sanitiser Unit (SU) and a secure Delivery Box (DB) for contactless medication delivery
using face recognition and the GLS. The development of the drone GLS system consists
of two microcontroller sensor circuits that visualise the delivery area's lighting conditions
to improve efficiency. In addition to the SU and DB, the drone GLS system also has a
Motion Detection Unit (MDU) and a Voice Command Guiding Unit (VCGU) for
medication delivery. The first development board is the Arduino Uno, which controls the
face recognition camera, DB, SU, MDU, and VCGU units. The second development
board is the ESP32 DevKitC V4, which controls the GLS. An Internet of Things (IoT)
microcontroller is connected to the Internet using IoT mobile apps. GLS is combined with
four light-dependent resistors and a light-intensity sensor. Those sensors visualise the
light conditions to enhance face recognition, and a GY-30 light intensity sensor is used
to measure the value of the illumination. The drone Facial Recognition Camera (FRC)
and GLS system has been tested on 5001 animal faces, 5030 static human photos inclusive
of 5000 photos from Flickr-Faces-HQ, 30 actual photos, and 35 real human volunteers'
faces. The overall accuracy of the face recognition system is 98.53%. The GLS has
enhanced the detection distance to almost double and increased the detection distance to
1.47 metres. A strong correlation was found between face recognition distance, light
direction, illumination, and light colour temperature (p<0.05). The drone GLS system can
detect real human face recognition with high accuracy, and the development is for the
purpose of performing medication delivery outdoors. |
|---|