Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Network
Quality inspection applications in industry are required to move towards a zero-defect manufacturing scenario, with non-destructive inspection and traceability of 100% of produced parts. Developing robust fault detection and classification models from the start-up of the lines is challenging due to...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/13/4361 |
_version_ | 1797528753276977152 |
---|---|
author | Julen Balzategui Luka Eciolaza Daniel Maestro-Watson |
author_facet | Julen Balzategui Luka Eciolaza Daniel Maestro-Watson |
author_sort | Julen Balzategui |
collection | DOAJ |
description | Quality inspection applications in industry are required to move towards a zero-defect manufacturing scenario, with non-destructive inspection and traceability of 100% of produced parts. Developing robust fault detection and classification models from the start-up of the lines is challenging due to the difficulty in getting enough representative samples of the faulty patterns and the need to manually label them. This work presents a methodology to develop a robust inspection system, targeting these peculiarities, in the context of solar cell manufacturing. The methodology is divided into two phases: In the first phase, an anomaly detection model based on a Generative Adversarial Network (GAN) is employed. This model enables the detection and localization of anomalous patterns within the solar cells from the beginning, using only non-defective samples for training and without any manual labeling involved. In a second stage, as defective samples arise, the detected anomalies will be used as automatically generated annotations for the supervised training of a Fully Convolutional Network that is capable of detecting multiple types of faults. The experimental results using 1873 Electroluminescence (EL) images of monocrystalline cells show that (a) the anomaly detection scheme can be used to start detecting features with very little available data, (b) the anomaly detection may serve as automatic labeling in order to train a supervised model, and (c) segmentation and classification results of supervised models trained with automatic labels are comparable to the ones obtained from the models trained with manual labels. |
first_indexed | 2024-03-10T10:03:01Z |
format | Article |
id | doaj.art-cef1816d426d4e08a123ab7a15301bb7 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T10:03:01Z |
publishDate | 2021-06-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-cef1816d426d4e08a123ab7a15301bb72023-11-22T01:47:54ZengMDPI AGSensors1424-82202021-06-012113436110.3390/s21134361Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial NetworkJulen Balzategui0Luka Eciolaza1Daniel Maestro-Watson2Electronics and Computer Science Department, Mondragon University, 20500 Arrasate, SpainElectronics and Computer Science Department, Mondragon University, 20500 Arrasate, SpainElectronics and Computer Science Department, Mondragon University, 20500 Arrasate, SpainQuality inspection applications in industry are required to move towards a zero-defect manufacturing scenario, with non-destructive inspection and traceability of 100% of produced parts. Developing robust fault detection and classification models from the start-up of the lines is challenging due to the difficulty in getting enough representative samples of the faulty patterns and the need to manually label them. This work presents a methodology to develop a robust inspection system, targeting these peculiarities, in the context of solar cell manufacturing. The methodology is divided into two phases: In the first phase, an anomaly detection model based on a Generative Adversarial Network (GAN) is employed. This model enables the detection and localization of anomalous patterns within the solar cells from the beginning, using only non-defective samples for training and without any manual labeling involved. In a second stage, as defective samples arise, the detected anomalies will be used as automatically generated annotations for the supervised training of a Fully Convolutional Network that is capable of detecting multiple types of faults. The experimental results using 1873 Electroluminescence (EL) images of monocrystalline cells show that (a) the anomaly detection scheme can be used to start detecting features with very little available data, (b) the anomaly detection may serve as automatic labeling in order to train a supervised model, and (c) segmentation and classification results of supervised models trained with automatic labels are comparable to the ones obtained from the models trained with manual labels.https://www.mdpi.com/1424-8220/21/13/4361anomaly detectionelectroluminescencesolar cellsneural networks |
spellingShingle | Julen Balzategui Luka Eciolaza Daniel Maestro-Watson Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Network Sensors anomaly detection electroluminescence solar cells neural networks |
title | Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Network |
title_full | Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Network |
title_fullStr | Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Network |
title_full_unstemmed | Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Network |
title_short | Anomaly Detection and Automatic Labeling for Solar Cell Quality Inspection Based on Generative Adversarial Network |
title_sort | anomaly detection and automatic labeling for solar cell quality inspection based on generative adversarial network |
topic | anomaly detection electroluminescence solar cells neural networks |
url | https://www.mdpi.com/1424-8220/21/13/4361 |
work_keys_str_mv | AT julenbalzategui anomalydetectionandautomaticlabelingforsolarcellqualityinspectionbasedongenerativeadversarialnetwork AT lukaeciolaza anomalydetectionandautomaticlabelingforsolarcellqualityinspectionbasedongenerativeadversarialnetwork AT danielmaestrowatson anomalydetectionandautomaticlabelingforsolarcellqualityinspectionbasedongenerativeadversarialnetwork |