Skripsi

PEWARNAAN CITRA GRAYSCALE MENGGUNAKAN GENERATIVE ADVERSARIAL NETWORKS (GANs)

---

Penilaian

---

Automatic grayscale image colorization is a challenge in computer vision that aims to transform black-and-white images into realistic color images. This study develops a colorization model based on generative adversarial networks using the Pix2Pix architecture, with U-Net as the generator and PatchGAN as the discriminator. Training was conducted in the CIELAB color space with three different learning rate configurations. The best results were obtained in configuration 3 (learning rate 0.000001) with an SSIM of 0.8909 and an MAE of 0.0762. Analysis shows that green and blue colors increasingly resemble the ground truth, while red still exhibits slight intensity deviations. The model effectively colorizes distinct objects but struggles with colors of high contrast.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2507001674	T169388	T1693882025	Central Library (Reference)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1693882025

Publisher

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

Collation

xii, 112 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

005.507

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Program aplikasi

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment