Text

PENGENALAN BAHASA ISYARAT INDONESIA (BISINDO) MENGGUNAKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK

---

Penilaian

---

Humans are social beings who need a media of interaction to communicate. One of the communication media is sign language represented in the form of hand gesture and commonly used as a media of communication for Deaf. Recognition of sign language is indispensable to facilitate communication between Deaf and non-Deaf. However, research on hand gesture recognition using Indonesian Sign Language (BISINDO) is still limited. To overcome this, the study will use convolutional neural network (CNN) to recognize 26 letters and numbers from 0 to 10 according to BISINDO. The data used is primary data with a total of 39455 data taken from 10 respondent with bright and dim lighting conditions as well as first-person and second-person viewpoints. From the results of CNN training that has been conducted, obtained the most optimal results with a training loss value of 0.0201 and validation loss of 0.0785 and training accuracy of 0.9948 and validation accuracy of 0.9839. Based on the test results using test data obtained accuracy of 98.3%, precision of 98.3%, recall of 98.4% and F1-Score of 99.3%. The proposed model is able to recognize BISINDO hand gestures well in both dim and bright lighting and from the point of view of both the first person and the second person.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2107002831	T52699	T526992021	Central Library (REFERENCES)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T526992021

Publisher

Inderalaya : Prodi Teknik Elektro, Fakultas Teknik Universitas Sriwijaya., 2021

Collation

xv, 169 hlm,: tab,: ilus,: 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

621.382 07

Content Type

-

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Teknik Komunikasi
Prodi Teknik Elektro

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment