Anomaly detection and automatic labeling for solar cell quality inspection based on Generative Adversarial Network

TL;DR

This paper introduces a two-phase system for solar cell quality inspection that uses GAN-based anomaly detection for initial fault identification and automatic labeling to train a supervised fault classification model, reducing manual effort.

Contribution

It presents a novel methodology combining GAN-based anomaly detection with automatic labeling to improve fault detection in solar cells with minimal manual annotation.

Findings

Anomaly detection can be effectively performed with limited data.

Automatic labels enable training supervised models comparable to manual labels.

The approach improves fault detection efficiency in solar cell manufacturing.

Abstract

Quality inspection applications in industry are required to move towards a zero-defect manufacturing scenario, withnon-destructive inspection and traceability of 100 % of produced parts. Developing robust fault detection and classification modelsfrom the start-up of the lines is challenging due to the difficulty in getting enough representative samples of the faulty patternsand the need to manually label them. This work presents a methodology to develop a robust inspection system, targeting thesepeculiarities, in the context of solar cell manufacturing. The methodology is divided into two phases: In the first phase, an anomalydetection model based on a Generative Adversarial Network (GAN) is employed. This model enables the detection and localizationof anomalous patterns within the solar cells from the beginning, using only non-defective samples for training and without anymanual labeling involved. In a second stage, as defective samples arise, the detected anomalies will be used as automaticallygenerated annotations for the supervised training of a Fully Convolutional Network that is capable of detecting multiple types offaults. The experimental results using 1873 EL images of monocrystalline cells show that (a) the anomaly detection scheme can beused to start detecting features with very little available data, (b) the anomaly detection may serve as automatic labeling in order totrain a supervised model, and (c) segmentation and classification results of supervised models trained with automatic labels arecomparable to the ones obtained from the models trained with manual labels.