2019
DOI: 10.5755/j01.eie.25.5.24357
|View full text |Cite
|
Sign up to set email alerts
|

Adaptation of Full-Reference Image Quality Assessment Methods for Automatic Visual Evaluation of the Surface Quality of 3D Prints

Abstract: Automatic visual quality assessment of the 3D printed surfaces is currently one of the most demanding challenges in additive manufacturing. Regardless of the applications of the computer vision for the 3D printing process monitoring purposes, a reliable surface quality evaluation during manufacturing may introduce brand new possibilities. The detection of some distortions and their automatic evaluation can be helpful when deciding to stop the process to save time, energy, and filament. In some cases, some furt… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2020
2020
2023
2023

Publication Types

Select...
5
2
1

Relationship

1
7

Authors

Journals

citations
Cited by 11 publications
(3 citation statements)
references
References 32 publications
(34 reference statements)
0
3
0
Order By: Relevance
“…A variety of sensing techniques are used for increasing road traffic safety. One approach is based on equipping vehicles [ 13 , 14 , 15 ] with new sensors and another by upgrading road infrastructure with traffic monitoring systems, from inductive loops to microwave radars and optical fibers [ 4 , 5 , 6 , 7 , 8 , 9 ]. Although many methods can be used for intelligent traffic systems, there are few key technologies widely accepted as reliable: surveillance camera systems, inductive loops and magnetic field sensors.…”
Section: Relate Workmentioning
confidence: 99%
“…A variety of sensing techniques are used for increasing road traffic safety. One approach is based on equipping vehicles [ 13 , 14 , 15 ] with new sensors and another by upgrading road infrastructure with traffic monitoring systems, from inductive loops to microwave radars and optical fibers [ 4 , 5 , 6 , 7 , 8 , 9 ]. Although many methods can be used for intelligent traffic systems, there are few key technologies widely accepted as reliable: surveillance camera systems, inductive loops and magnetic field sensors.…”
Section: Relate Workmentioning
confidence: 99%
“…Image augmentation techniques are not used. ADAM [55] is used as an optimizer during the training with a learning rate 0.0001. As shown in Figure 8, the average segmentation accuracy surpasses 85% on the validation dataset in less than 15 epochs.…”
Section: Step 1: Training Of the Semantic Segmentation Modelmentioning
confidence: 99%
“…An additional problem might be related fo different colors and luminance of the acquired images and those rendered from the 3D models. Hence, the idea of the calculation of the average mutual similarity is proposed for the images of the 3D printed surfaces assuming the side location of cameras, initially investigated in the papers [54,55] exclusively for classification purposes. Such side-view mounting of a camera makes it possible to acquire images with visible individual layers of the filament, as illustrated in Figure 1.…”
Section: Proposed Approachmentioning
confidence: 99%