Aktivitäten pro Jahr
Abstract
Sorting mixed mail has become increasingly important for postal service
providers. Low-value consignments, that have unfavorable material
properties, such as odd shape or flexibility, have increased in terms of
volume in recent years and cause substantial problems in sorting
processes. In order to automate the sorting of mixed-mail items the so-
called ‘Free-Fall-Sorter’ was developed at the Institute of Logistics
Engineering at Graz University of Technology. In this paper, an image
classification system is proposed to pre-sort mixed mail for this new
machine. The computer vision system used for the experiments applies
transfer learning to a publicly available convolutional neural network
called ‘MobileNet V2’. The classification system is designed to be
composed of low-cost hardware components, yet capable of providing
real-time inference. The results show a high accuracy of the re-trained
model.
providers. Low-value consignments, that have unfavorable material
properties, such as odd shape or flexibility, have increased in terms of
volume in recent years and cause substantial problems in sorting
processes. In order to automate the sorting of mixed-mail items the so-
called ‘Free-Fall-Sorter’ was developed at the Institute of Logistics
Engineering at Graz University of Technology. In this paper, an image
classification system is proposed to pre-sort mixed mail for this new
machine. The computer vision system used for the experiments applies
transfer learning to a publicly available convolutional neural network
called ‘MobileNet V2’. The classification system is designed to be
composed of low-cost hardware components, yet capable of providing
real-time inference. The results show a high accuracy of the re-trained
model.
Titel in Übersetzung | Ein Verfahren zur Vorsortierung von Postsendungen mittels faltenden neuronalen Netzen und Transferlernen |
---|---|
Originalsprache | englisch |
Titel | MHCL 2022 |
Untertitel | XXIV International Conference on "Material Handling, Constructions and Logistics" September 21st - 23rd, 2022, Belgrade |
Erscheinungsort | Belgrade |
Herausgeber (Verlag) | Faculty of Mechanical Engineering, Belgrade University |
Seiten | 71-80 |
Seitenumfang | 10 |
ISBN (Print) | 978-86-6060-134-8 |
Publikationsstatus | Veröffentlicht - Sept. 2022 |
Veranstaltung | 24th International Conference on Material Handling, Contructions and Logistics: MHCL 2022 - Faculty of mechanical Engineering - University Belgrade, Belgrad, Serbien Dauer: 21 Sept. 2022 → 23 Sept. 2022 https://www.mhcl.info |
Konferenz
Konferenz | 24th International Conference on Material Handling, Contructions and Logistics |
---|---|
Land/Gebiet | Serbien |
Ort | Belgrad |
Zeitraum | 21/09/22 → 23/09/22 |
Anderes | XXIV International Conference on Material Handling, Contructions and Logistics |
Internetadresse |
Schlagwörter
- polybag
- Convolutional neural networks (CNNs)
- image classification
- deep learning
- low-value consignment
Fields of Expertise
- Mobility & Production
Aktivitäten
- 1 Vortrag bei Konferenz oder Fachtagung
-
A method for pre-sorting mixed mail using convolutional neural networks and transfer learning
Michael Schadler (Redner/in)
21 Sept. 2022 → 23 Sept. 2022Aktivität: Vortrag oder Präsentation › Vortrag bei Konferenz oder Fachtagung › Science to science