Workspace Classification Model for Manufacturing Using Deep Learning-Based Scene Recognition 


Vol. 47,  No. 6, pp. 855-861, Jun.  2022
10.7840/kics.2022.47.6.855


PDF
  Abstract

With the recent development of machine learning technology, various methods of scene recognition through deep learning are being studied in industrial fields. The performance of scene recognition using deep learning is greatly affected by the algorithm structure and learning method. As a result of the preliminary study for this study, it was confirmed that the manufacturing workspace classification accuracy by the data model trained with the Places365 dataset was 54%, which was difficult to use in the real environment. Therefore, in this paper, in order to solve this problem, a classification model of the manufacturing workspace is presented using a scene recognition technique after learning by reconstructing a learning dataset with three strategies based on CNN (Convolutional Neural Network). When learning through the proposed algorithm and data set configuration method, it was confirmed that the scene recognition performance for the manufacturing workspace was improved by 28%p compared to the existing method.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

J. S. Kim and D. M. Lee, "Workspace Classification Model for Manufacturing Using Deep Learning-Based Scene Recognition," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 6, pp. 855-861, 2022. DOI: 10.7840/kics.2022.47.6.855.

[ACM Style]

Jeong Su Kim and Dong Myung Lee. 2022. Workspace Classification Model for Manufacturing Using Deep Learning-Based Scene Recognition. The Journal of Korean Institute of Communications and Information Sciences, 47, 6, (2022), 855-861. DOI: 10.7840/kics.2022.47.6.855.

[KICS Style]

Jeong Su Kim and Dong Myung Lee, "Workspace Classification Model for Manufacturing Using Deep Learning-Based Scene Recognition," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 6, pp. 855-861, 6. 2022. (https://doi.org/10.7840/kics.2022.47.6.855)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880