Artificial intelligence

Scientific journal

ISSN 2710-1673

ONLINE: ISSN 2710-1681

Select your language


Methods of artificial intelligence in the software management system for critical sizes control equipment on the basis of computer vision

Voronov А.1, Ganchenko V.2, Doudkin A.3, Аvakaw S.4, Dedkov A.5, Sholomicki V.6
1 United Institute of Informatics Problems of NAS of Belarus
2 United Institute of Informatics Problems of the National Academy of Sciences of Belarus
3 United Institute of Informatics Problems of the National Academy of Sciences of Belarus
4 KBTEM-OMO
5 KBTEM-OMO
6 KBTEM-OMO

Full text (PDF)

UDC: 004.93'1
Publication Language: Russian
Stuc. intelekt. 2018; 23(4):39-45

Abstract: In this paper the functional requirements and the structure of software control system for equipment of critical sizes inspection on the basis of computer vision are described. The artificial intelligence methods including Network-in-Network type neural networks and deep learning were applied for solving problems of analyzing images. The advantages of the architecture of Network-in-Network are considered. Application of Network-in- Network technology allows identifying effectively defects that is especially important for software engineering for equipment of critical sizes inspection of VLSI manufacturing based on submicron technology.

Keywords: neural networks, VLSI, NiN, Computer Vision

References:

  1. Convolutional Neural Networks for VisualRecognition(2015)[Online].Available:http://cs231n.github.io/convolutional-networks/
  2. Network In Network (1991) [Online]. Available:https://arxiv.org/pdf/1312. 4400v3.pdf.
  3. Network In Network architecture: The beginningof Inception (2017) [Online]. Available:http://teleported.in/posts/network-in-network.
  4. An Analysis of Deep Neural Network Models ForPractical Applications (1991) [Online]. Available:https://arxiv.org/pdf/1605 07678v4.pdf.
  5. Dostanko, A.P., Avakov, S.M., Ageev, O.A.,Batura, M.P. (2016) Tehnologicheskie komplek-syiintegrirovannyih protsessov proizvodstva izdeliyelektroniki. Minsk: Belaruskaya navuka. – 251 s.
  6. Nikolenko, S., Kadurin, A., Arhangelskaya, E.(2018) Glubokoe obuchenie. SPb.: Piter, 480 s.
  7. Tensorflow API documentation [Online].Available: /api_docs/python/tf/nn/softmax_cross_entropy_with_logits_v2.

View full text (PDF)