Skripsi

KLASIFIKASI PEROKOK BERDASARKAN KARAKTERISTIK FISIK WAJAH MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK

---

Indonesia has a high prevalence of active smokers, reaching 70.6% among adult males in 2019. Smoking is known to cause various physical changes in facial skin, such as accelerated aging and wrinkles. This research aims to develop and test an automated system for classifying smokers based on facial physical characteristics using a deep learning approach. The method used is Transfer Learning, utilizing the FaceNet model as a feature extractor. Face detection is performed using Multi-task Cascaded Convolutional Networks (MTCNN). The dataset is a combination of primary and secondary data, totaling approximately 1000 images. The model's performance is evaluated using the metrics of Accuracy, Loss, Confusion Matrix, Precision, Recall, and F-1 Score. The results show that the system was successfully developed with a Validation Accuracy of 74,68%, Precision of 0,729, Recall of 0,729, and F-1 Score of 0,729. Qualitative analysis using Grad-CAM demonstrates that the model tends to focus on facial areas relevant to the clinical criteria for a smoker's face, such as the mouth, under-eye bags, and cheeks. This research contributes by proving the feasibility of using CNN for classifying physical traits of a smoker’s face.

---

Availability

Detail Information

Series Title

-

Call Number

T1896722025

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xv, 83 hlm.; ilus.; tab.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.320 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Convolutional Neural Network

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment