EUVIMEDEUVIMED entdecken
Crossref

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

HO Sy Phuong, PHAN Van Du, LE Van Chuong, TA Hung Cuong

Vinh University Journal of Science · 2018

DOI: 10.56824/vujs.2018tn26 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Vinh University Journal of Science, 2018. Verfasst von HO Sy Phuong und 3 weiteren Autor:innen. Die Arbeit nennt 17 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

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.

Zitieren

APA

Phuong, H. S., Du, P. V., Chuong, L. V., & Cuong, T. H. (2018). Applying neural networks, convolutional neural networks and combination of CPUs and GPUs to increase calculating performance for image classification. Vinh University Journal of Science. https://doi.org/10.56824/vujs.2018tn26

Vancouver

Phuong HS, Du PV, Chuong LV, Cuong TH. Applying neural networks, convolutional neural networks and combination of CPUs and GPUs to increase calculating performance for image classification. Vinh University Journal of Science. 2018. doi:10.56824/vujs.2018tn26

BibTeX
@article{Phuong2018,
  author = {Phuong, HO Sy and Du, PHAN Van and Chuong, LE Van and Cuong, TA Hung},
  title = {Applying neural networks, convolutional neural networks and combination of CPUs and GPUs to increase calculating performance for image classification},
  journal = {Vinh University Journal of Science},
  year = {2018},
  doi = {10.56824/vujs.2018tn26}
}
Kostenlos starten →