Facial Recognition System for Car Ignition using OpenCV, DLIB Framework and Raspberry Pi 3 Model B / John Paul G. Abedante.
Material type:
TextPublisher: Rosario, Cavite : Cavite State University-CCAT Campus, 2018Description: xiii, 58 leaves : illustrations ; 28 cmSubject(s): Human face recognition | Facial recognition | Amperage draw | Face recognition algorithm | DLIB framework | Raspberry Pi | Face detection | Car ignitionLOC classification: UM TK 7870 | E88 2017Summary: ABENDANTE, JOHN PAUL G. Facial Recognition System for Car Ignition using OpenCV, DLIB Framework and Raspberry Pi 3 Model B. Design Project. Department of Engineering. Cavite State University-Cavite College of Arts and Trades Campus, Rosario, Cavite. June 2018. Adviser: Mrs. Diane P. Arayata. Technical adviser: Fernando M. Cielo.
The study was conducted from May 2017 to March 2018 to create a system to ignite a car engine using facial recognition. Specifically, the study aimed to: 1) develop a device that will manage the car’s ignition without the use of a key; 2) evaluate the speed of the system’s initialization; 3) conduct an accuracy test for the face recognition system; 4) test the device for its durability ; and 5) measure the amperage draw of the device.
The device was tested by involving 6 participants that were divided into 2 categories the registered and the unregistered individuals. Each subject has undergone 3 consecutive trials.
The results of the evaluation revealed that the system’s face recognition algorithm used was reliable, but was considered inefficient due to the high initialization time which is about 1 minute. The accuracy is affected by different poses and illumination.
| Item type | Current location | Collection | Shelving location | Call number | Copy number | Status | Date due | Barcode |
|---|---|---|---|---|---|---|---|---|
Thesis/Manuscripts/Dissertations
|
Cavite State University - CCAT Campus | Thesis/Manuscript/Dissertation | TH | UM TA 1637 A24 2018 (Browse shelf) | 1 copy | Available | T0004201 |
Browsing Cavite State University - CCAT Campus shelves, Shelving location: TH, Collection: Thesis/Manuscript/Dissertation Close shelf browser
|
|
|
|
|
|
|
|
||
| UM TA 1634 D48 2013 Automated Baggage Counter / | UM TA 1634 G37 2019 Automated Fish Sorting and Harvesting System for Nile Tilapia (Oreochromis Niloticus) Using Computer Vision / | UM TA 1637 A24 2018 Facial Recognition System for Car Ignition using OpenCV, DLIB Framework and Raspberry Pi 3 Model B / | UM TA 1637 A24 2018 Facial Recognition System for Car Ignition using OpenCV, DLIB Framework and Raspberry Pi 3 Model B / | UM TA 1637 A43 2019 Coffee Berry Grader and Ripeness Detector / | UM TA 1637 C66 2019 Real Time Obstacle Detection for Visual Impaired Individuals / | UM TA 1637 G83 2018 Development of PEREA: A Raspberry Pi-Based Personal Reading Assistant for the Blind Persons / |
Design Project (BSCpE)--Cavite State University-CCAT Campus, 2018.
Includes bibliographical references and appendices.
ABENDANTE, JOHN PAUL G. Facial Recognition System for Car Ignition using OpenCV, DLIB Framework and Raspberry Pi 3 Model B. Design Project. Department of Engineering. Cavite State University-Cavite College of Arts and Trades Campus, Rosario, Cavite. June 2018. Adviser: Mrs. Diane P. Arayata. Technical adviser: Fernando M. Cielo.
The study was conducted from May 2017 to March 2018 to create a system to ignite a car engine using facial recognition. Specifically, the study aimed to: 1) develop a device that will manage the car’s ignition without the use of a key; 2) evaluate the speed of the system’s initialization; 3) conduct an accuracy test for the face recognition system; 4) test the device for its durability ; and 5) measure the amperage draw of the device.
The device was tested by involving 6 participants that were divided into 2 categories the registered and the unregistered individuals. Each subject has undergone 3 consecutive trials.
The results of the evaluation revealed that the system’s face recognition algorithm used was reliable, but was considered inefficient due to the high initialization time which is about 1 minute. The accuracy is affected by different poses and illumination.
There are no comments on this title.