Computer Vision in Industry

Izdanje: International Scientific Conference on Information Technology and Data Related Research

DOI: 10.15308/Sinteza-2019-439-444

Oblast: Data Science & Digital Broadcasting Systems

Stranice: 439-444

Apstrakt:
This paper presents a system of computer vision designed to be used in industry for product classification. The system is based on the TensorFlow and Keras software libraries and the Python programming language. In the production process, uniform cylindrical parts reach the conveyor designed so that each part can be in one of two possible positions. The computer vi- sion system detects the exact position of the part, or its orientation, and this information is further used during the transportation process. The purpose of the procedure is for all parts to take the same orientation in relation to the production line.
Ključne reči: industry, computer vision, machine learning, Keras, TensorFlow, Python, RaspberryPi
Priložene datoteke:
  • 439-444 ( veličina: 1,55 MB, broj pregleda: 240 )

Preuzimanje citata:

BibTeX format
@article{article,
  author  = {D. Đurić and V. Matić}, 
  title   = {Computer Vision in Industry},
  journal = {International Scientific Conference on Information Technology and Data Related Research},
  year    = 2020,
  pages   = {439-444},
  doi     = {10.15308/Sinteza-2019-439-444}
}
RefWorks Tagged format
RT Conference Proceedings
A1 Dobrivoje Đurić
A1 Vladimir Matić
T1 Computer Vision in Industry
AD Univerzitet Singidunum, Beograd, Beograd, Srbija
YR 2020
NO doi: 10.15308/Sinteza-2019-439-444
Unapred formatirani prikaz citata
D. Đurić and V. Matić, Computer Vision in Industry, Univerzitet Singidunum, Beograd, 2020, doi:10.15308/Sinteza-2019-439-444