Pratama, Rendi and Kurniawan, Rio (2023) PENERAPAN METODE CONVOLUTION NEURAL NETWORK MENGGUNAKAN WEBCAM UNTUK MENGANALISIS EKSPRESI WAJAH SISWA YANG BERMASALAH PADA UNIT BIMBINGAN KONSELING (STUDI KASUS DI SMAN 1 PENENGAHAN LAMPUNG SELATAN). Skripsi thesis, Institut Informatika dan Bisnis Darmajaya.
![[img]](http://repo.darmajaya.ac.id/style/images/fileicons/text.png) |
Text
1. Cover.pdf Download (144kB) |
|
Text
2. Abstract.pdf Download (14kB) |
|
Text
3. Halaman Persetujuan dan Pengesahan.pdf Download (263kB) |
|
Text
4. Daftar Isi.pdf Download (197kB) |
|
Text
5. BAB I.pdf Download (126kB) |
|
Text
6. BAB II.pdf Download (512kB) |
|
Text
7. BAB III.pdf Download (153kB) |
|
Text
8. BAB IV.pdf Download (927kB) |
|
Text
9. BAB V.pdf Download (11kB) |
|
Text
10. Daftar Pustaka.pdf Download (229kB) |
|
Text
11. Lampiran.pdf Download (796kB) |
Abstract
In the context of Guidance and Counseling (BK), the importance of detecting students' facial expressions is one of the main problems that need to be examined. Expression detection becomes one of the main problems that need to be examined. Students often face difficulties in expressing their feelings verbally, especially when experiencing emotions such as depression, anxiety, or stress. Therefore, facial expression detection methods were the main focus of this research. This research focused on the use of Convolution Neural Network as a method to classify students' facial expressions. Convolution Neural Network is a technique that can identify objects from colors and contours in an image, which can be used to understand facial expressions in an image, which can be used to understand students' facial expressions in a counseling guidance context. The research data consisted of 618 student images which were divided into 7 expression classes. The preprocessing process involved image resizing, conversion to grayscale, and label replacement class. A Convolution Neural Network model was developed with various layers, including Conv2D, MaxPooling2D, Conv2D, Flatten Layer, Dense Layer, Dropout Layer, and Dense Layer after going through the training process, the model showed an improvement in the accuracy of the training data, although the results of testing the test data varied. The total accuracy of the model during testing was about 33%, which was then validated by testing the training data. Then validated the model by testing the model on 30 new test data which resulted in an accuracy of 20%. Although this accuracy still needs to be improved, this research provides a solid foundation for further development in student facial expression detection during counseling sessions. Future research can focus on improving the accuracy of the model by increasing the amount of data and designing a more complex Convolution Neural Network architecture.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Subjects: | Ilmu Komputer eSkripsi |
| Divisions: | Skripsi/TA & PKPM/KP - Fakultas Ilmu Komputer > Teknik Informatika |
| Depositing User: | Rendi Pratama |
| Date Deposited: | 03 Jun 2024 01:23 |
| Last Modified: | 03 Jun 2024 01:23 |
| URI: | http://repo.darmajaya.ac.id/id/eprint/16328 |
Actions (login required)
 |
View Item |