Difference between revisions of "ML-TN-003 — AI at the edge: visual inspection of assembled PCBs for defect detection — Part 3"

From DAVE Developer's Wiki
Jump to: navigation, search
Line 21: Line 21:
 
==Building the dataset==
 
==Building the dataset==
 
===Defects generation and acquisition===
 
===Defects generation and acquisition===
 +
{| class="wikitable" style="text-align:center;"
 +
|- style="font-weight:bold;"
 +
|+'''Anomalies, generation process and numerosity'''
 +
! Anomaly
 +
! P&P
 +
! Serigraphy
 +
! Manual
 +
! Numerosity
 +
|-
 +
| Missing
 +
| ✓
 +
|
 +
|
 +
| 21
 +
|-
 +
| Manhattan
 +
|
 +
|
 +
| ✓
 +
| 4
 +
|-
 +
| Shift x-axis
 +
| ✓
 +
|
 +
|
 +
| 51
 +
|-
 +
| Shift y-axis
 +
| ✓
 +
|
 +
|
 +
| 58
 +
|-
 +
| Shift&Rotation<br />(shift x+z-axes)
 +
| ✓
 +
|
 +
|
 +
| 57
 +
|-
 +
| Rotation<br />(z-axis)
 +
| ✓<br />
 +
|
 +
|
 +
| 55
 +
|-
 +
| Under soldering
 +
|
 +
| ✓
 +
|
 +
| All
 +
|-
 +
| Over soldering
 +
|
 +
| ✓
 +
|
 +
| All
 +
|}
 +
 
===Class subdivision and labelling===
 
===Class subdivision and labelling===
 
===Soldering regions extraction===
 
===Soldering regions extraction===
Line 90: Line 148:
 
*Tero Karras, Timo Aila, Samuli Laine, Jaakko Lehtinen, [https://arxiv.org/abs/1710.10196 ''Progressive Growing of GANs for Improved Quality, Stability, and Variation''], February 2018.
 
*Tero Karras, Timo Aila, Samuli Laine, Jaakko Lehtinen, [https://arxiv.org/abs/1710.10196 ''Progressive Growing of GANs for Improved Quality, Stability, and Variation''], February 2018.
 
* Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, Aaron Courville, [https://arxiv.org/abs/1704.00028 ''Improved Training of Wasserstein GANs''], December 2017.
 
* Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, Aaron Courville, [https://arxiv.org/abs/1704.00028 ''Improved Training of Wasserstein GANs''], December 2017.
 +
*[https://shop.ipc.org/IPC-A-610E-English-D ''IPC-A-610E: Acceptability of Electronic Assemblies''], a standard developed by IPC, April 2010.

Revision as of 12:27, 14 April 2021

HOME SOMs SBCs ToloMEO Embedded Assistant Sale-tag.pngGET A QUOTE Customer-service.pngONLINE HELPDESK
Roadmap IoT Services ML/AI services Embedded Design Services
Info Box
NeuralNetwork.png Applies to Machine Learning


History[edit | edit source]

Version Date Notes
1.0.0 March 2021 First public release

Introduction[edit | edit source]

Building the dataset[edit | edit source]

Defects generation and acquisition[edit | edit source]

Anomalies, generation process and numerosity
Anomaly P&P Serigraphy Manual Numerosity
Missing ✓ 21
Manhattan ✓ 4
Shift x-axis ✓ 51
Shift y-axis ✓ 58
Shift&Rotation
(shift x+z-axes) 		✓ 57
Rotation
(z-axis) 		✓
 55
Under soldering ✓ All
Over soldering ✓ All

Class subdivision and labelling[edit | edit source]

Soldering regions extraction[edit | edit source]

Data augmentation with image synthesis[edit | edit source]

Generative adversarial networks[edit | edit source]

Progressive GAN implementation[edit | edit source]

Results validation[edit | edit source]

Class Synth image Resolution
(pixel) 		Google Colab
(min) 		AWS SageMaker
(min)

missing 		full 512 × 512 ~480 ~410
upper/lower 256 × 256 ~435 ~310

tombstoning 		full 512 × 512 ~460 ~390
upper/lower 256 × 256 ~420 ~300
Synthesized images for missing class
Synthesized images for tombstoning class
T-SNE algorithm results for missing class synthesized images
512 × 512 resolution full images
256 × 256 resolution full images
256 × 256 resolution upper and lower region images


T-SNE algorithm results for tombstoning class synthesized images
512 × 512 resolution full images
256 × 256 resolution full images
256 × 256 resolution upper and lower region images


Useful links[edit | edit source]

HOME SOMs SBCs ToloMEO Embedded Assistant Sale-tag.pngGET A QUOTE Customer-service.pngONLINE HELPDESK
Roadmap IoT Services ML/AI services Embedded Design Services
Retrieved from "https://wiki.dave.eu/index.php?title=ML-TN-003_—_AI_at_the_edge:_visual_inspection_of_assembled_PCBs_for_defect_detection_—_Part_3&oldid=13495"