magistrsko delo
Abstract
Delo opisuje nekaj najsodobnejših pristopov reševanja inženirskih problemov z uporabo globokega učenja in predstavlja sistem za zaznavanje okolice v dinamičnem proizvodnem okolju. Algoritmi strojnega učenja ponujajo v kombinaciji z optičnimi senzorji (kamerami) možnost reševanja izjemno kompleksnih problemov, katerim so do sedaj bili kos le ljudje. Avtomatizacija procesov, pretok informacij med stroji in ljudmi ter pametna analiza podatkov s procesiranjem v oblaku, so le nekateri izzivi, ki jih naslavlja Industrija 4.0. Magistrsko delo predstavlja dinamičen sistem strojnega vida, ki ponuja rešitev na področju klasifikacije in lokalizacije poljubnih objektov v proizvodnih sistemih.
Keywords
proizvodni sistemi;strojni vid;globoko učenje;industrija 4.0;magistrske naloge;
Data
Language: |
Slovenian |
Year of publishing: |
2019 |
Typology: |
2.09 - Master's Thesis |
Organization: |
UM FS - Faculty of Mechanical Engineering |
Publisher: |
[J. Hernavs] |
UDC: |
[004.85+004.93]:681.586.5(043.2) |
COBISS: |
22352150
|
Views: |
891 |
Downloads: |
191 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
Using deep learning and machine vision for object recognition in manufacturing systems |
Secondary abstract: |
Paper depicts several contemporary Deep Learning approaches, as well as an object recognition system for applications in a field of manufacture engineering. Until now, there were a lot of specific tasks that only human could manage. Opportunity for complex problem solving presents itself in the combination of optical instruments and machine learning algorithms. Process automation, data exchange and elaborate analysis systems with Cloud computing are just a few examples of the challenges, addressed by the Industry 4.0. This work presents a dynamic system of machine vision that offers object classification and localization. |
Secondary keywords: |
manufacturing systems;machine vision;deep learning;Industry 4.0; |
URN: |
URN:SI:UM: |
Type (COBISS): |
Master's thesis/paper |
Thesis comment: |
Univ. v Mariboru, Fak. za strojništvo, Računalniško inženirsko modeliranje |
Pages: |
IV, 49 f. |
ID: |
11023264 |