CD Skripsi

Klasifikasi Huruf Hijaiyah Dalam Bahasa Isyarat Menggunakan Metode Cnn

---

ABSTRACK
Inclusive education, particularly in the learning of Hijaiyah letters (Arabic
alphabet), remains a significant challenge for deaf students. Conventional media
often fails to bridge this communication barrier, a problem exacerbated by the
difficulty of distinguishing between several very similar hand gestures, a real issue
observed at SLB Panam Mulia Pekanbaru. To address this gap and promote
educational equity, this research develops an innovative solution: a system capable
of classifying Hijaiyah letter hand gestures using artificial intelligence, specifically
a Convolutional Neural Network (CNN).The main objective of this study is to create
an image-based system that is more effective and accurate, surpassing the
performance of previous research. The authors meticulously optimized the CNN
model, utilizing the Batch Normalization technique to stabilize the learning
process, and implementing a very low Learning Rate (0.0001) to ensure the model
learns smoothly and accurately.The results are highly encouraging. The developed
system successfully achieved a Validation Accuracy of 88.78%. This figure
represents a significant improvement, far exceeding the previous reference study's
accuracy of 76.92%. Although there are still minor challenges in distinguishing
gestures that are truly identical, such as the letters ain and ghain, the high accuracy
achieved proves that this model is a very robust and effective technological solution
for supporting and improving the quality of inclusive Hijaiyah education for the
deaf and hard-of-hearing.
Keywords : Sign Language, CNN, Classification, Hijaiyah

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

1907113855

Penerbit

Pekanbaru : Universitas Riau – F.TEKNIK – INFORMATIKA., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

1907113855

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

TEKNIK INFORMATIKA

Info Detail Spesifik

-

Pernyataan Tanggungjawab

daus

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar