3A

, Applying neural networks, convolutional neural networks and combination of CPUs and GPUs to increase calculating performance for image classification

Authors:
Hồ Sỹ Phương, Phan Văn Dư, Lê Văn Chương, Tạ Hùng Cường
View:
0
Position:
0/0
This paper presents and compares the image classification methods based on MLPs and CNNs. Training data is 500,000 pictures of 10 different objects. The first architecture to be used is MLPs network that contains 3,853,298 weights, the second architecture is CNNs with 528,054 weights. This paper proposes several methods and architectures network to avoid overfitting phenomenon and increases the accuracy of modeling approximately 80%. Besides on it, the paper also presents and compares time training of models using CPUs, and combining CPUs with GPUs.
Relate
Keywords

Vinh University journal of science

Tạp chí khoa học Trường Đại học Vinh

ISSN: 1859 - 2228

Governing body: Vinh University

  • Address: 182 Le Duan - Vinh City - Nghe An province
  • Phone: (+84) 238.3855.452 - Fax: (+84) 238.3855.269
  • Email: vinhuni@vinhuni.edu.vn
  • Website: https://vinhuni.edu.vn

 

License: 163/GP-BTTTT issued by the Minister of Information and Communications on May 10, 2023

Open Access License: Creative Commons CC BY NC 4.0

 

CONTACT

Editor-in-Chief: Assoc. Prof., Dr. Tran Ba Tien
Email: tientb@vinhuni.edu.vn

Deputy editor-in-chief: Dr. Phan Van Tien
Email: vantienkxd@vinhuni.edu.vn

Sub-Editor: Dr. Do Mai Trang
Email: domaitrang@vinhuni.edu.vn

Editorial assistant: Msc. Le Tuan Dung, Msc. Phan The Hoa, Msc. Pham Thi Quynh Nga, Msc. Tran Thi Thai

  • Address: 4th Floor, Executive Building, No. 182, Le Duan street, Vinh city, Nghe An province.
  • Phone: (+84) 238-385-6700 | Hotline: (+84) 97-385-6700
  • Email: editors@vujs.vn
  • Website: https://vujs.vn

img