Building a Computer Vision Finger Detection by Displaying Numbers with Pytorch
DOI:
https://doi.org/10.55927/marcopolo.v3i5.43Keywords:
CNN, Deep Learning, Pytorch, Finger PatternAbstract
This study aims to build a computer vision system that is able to detect fingers and display numbers based on recognized finger patterns. This technology is designed using the PyTorch deep learning framework to process images in real-time and identify the number of fingers displayed by the user. The process starts from image acquisition via camera, followed by image pre-processing, hand segmentation, feature extraction, to finger count classification using a deep learning model. The system is tested under various lighting conditions and backgrounds to measure its accuracy and reliability. The test results show that the model is able to recognize finger patterns with a high level of accuracy and respond quickly. This implementation has the potential to be applied in various applications such as touchless interfaces, communication aids, and gesture-based interactive systems.
References
A. Anggraini and H. Zakaria, "Penerapan Metode Deep Learning Pada Aplikasi Pembelajaran Menggunakan Sistem Isyarat Bahasa Indonesia Menggunakan Convolutional Neural Network (Studi Kasus: SLB-BC Mahardika Depok)," 2023.
Meirza and N. R. Puteri, "Implementasi Metode YOLOV5 dan Tesseract OCR untuk Deteksi Plat Nomor Kendaraan," 2024.
K. Azmi, S. Defit, and Sumijan, "Implementasi Convolutional Neural Network (CNN) Untuk Klasifikasi Batik Tanah Liat Sumatera Barat," 2023.
M. Tarmizi and Yahfizham, "Perspektif Mahasiswa Terhadap Penggunaan Kecerdasan Buatan ChatGPT dalam Penyusunan Tugas Akhir," 2024.
R. Rizal, Martanto, and Y. A. Wijaya, "Analisa Dataset Software Defined Network Intrusion Menggunakan Algoritma Deep Learning H2O," 2022.
