Face recognition employees attendance system
Face recognition uses a variety of technologies and locations to carry out the attendance system. In order to recognise a face in real-time settings utilising a specific purpose device, attendance systems require accurate results. Video architecture is also achieved in our design by piercing the cam...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/6983/1/24p%20MUNEF%20HASAN%20ABDULLAH%20AL%20NASSER.pdf http://eprints.uthm.edu.my/6983/2/MUNEF%20HASAN%20ABDULLAH%20AL%20NASSER%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/6983/3/MUNEF%20HASAN%20ABDULLAH%20AL%20NASSER%20WATERMARK.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Face recognition uses a variety of technologies and locations to carry out the attendance system. In order to recognise a face in real-time settings utilising a specific purpose device, attendance systems require accurate results. Video architecture is also achieved in our design by piercing the camera via a stoner- friendly interface. The Overeater (Histogram of Acquainted Grade) algorithm is used to recognise and segment the face from the VHS frame. Garbling a photo using the Overeater method to obtain a simplified interpretation of the image is the first phase, or pre-processing stage. Find the part of the image that most closely resembles a general Overeater encoding of a face using this simplified image. Also in the next step, figuring out the face's disguise by chancing the primary landmarks in the face. Once we've located those landmarks, we can utilise them to anchor the image such that the eyes and mouth are centred. Run the centred face image through a neural network that understands how to measure facial traits. Save those 128 measurements for later. Examine all of the faces we've measured in the past to find who has the most similar measurements to ours. That's the result of our match. Overall, we developed a Python programme that takes an image from a database and does all of the necessary changes for recognition, as well as checks the image in videos or in real time by accessing the camera using a Stoner-friendly interface. After a successful match is made, the name and time of the individual in attendance is recorded. |
|---|