Face recognition by artificial neural network using MATLAB
Facial Recognition considering one of the most difficult operations due to the unruly amount of datasets. However, human could easy to recognize an emotion while inconceivably for a computers. Artificial Neural Networks (ANN) provides an exceedingly smart solution in terms of recognition performance...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2021
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/42385/1/Face%20recognition%20by%20artificial%20neural%20network%20using%20MATLAB.pdf http://umpir.ump.edu.my/id/eprint/42385/2/Face%20recognition%20by%20artificial%20neural%20network%20using%20MATLAB_ABS.pdf http://umpir.ump.edu.my/id/eprint/42385/ https://doi.org/10.1109/ICCCA52192.2021.9666434 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Facial Recognition considering one of the most difficult operations due to the unruly amount of datasets. However, human could easy to recognize an emotion while inconceivably for a computers. Artificial Neural Networks (ANN) provides an exceedingly smart solution in terms of recognition performance when deal might as well with the data. In this paper human faces have been detected through artificial neural network using MATLAB simulation to find out the impression via recognizing the expression of the faces obtained from the database that containing 266 samples with various expressions within the wide ages. Consequently, many pre-classified datasets such as Japanese Female Facial Expression (JFFE), Face and Gesture Recognition (FG-NET), Face Expression Recognition Dataset 2013 (FER-2013), and Cohn Kanade Dataset (CK +) were studied to achieve a comprehensive model that could contribute the scientific research. The study investigated an obtained dataset to demonstrate the efficiency and solidarity of the proposed through to focus positively on the facial impression and its fluctuations. The result clearly shows that LEARN Gradient Descent with Momentum weight (LEARNGDM) is the best learning function to get an accomplishment with an average error equal to 0.01257, validation ratio 97.462, and 98.67232 precision. |
|---|