Split-Brain Autoencoder Approach for Surface Defect Detection

Ulutas, Tolga; Oz, Muhammed Ali Nur; Mercimek, Muharrem; Kaymakçı, Özgür Turay

doi:10.1109/icecce49384.2020.9179311

Cited by 5 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GANbased generative models [1,17,22,42,58] estimate anomalies through a discriminative network that compares the query image with randomly sampled samples from the latent space of the generative network. Besieds, there are several works that introduce proxy tasks based on the generative paradigm, such as image inpainting [44,60] and attribute prediction [34,49,57]. In addition, Normalizing Flow-based methods [19,38,39,59] are a combination of deep feature embeddings and generative models, which estimate accurate data likelihoods in the latent space by learning bijective transformations between normal sample distributions and specified densities.…”

Section: Related Workmentioning

confidence: 99%

DiffusionAD: Denoising Diffusion for Anomaly Detection

Zhang¹,

Wang²,

Wu³

et al. 2023

Preprint

View full text Add to dashboard Cite

Anomaly detection is widely applied due to its remarkable effectiveness and efficiency in meeting the needs of realworld industrial manufacturing. We introduce a new pipeline, DiffusionAD, to anomaly detection. We frame anomaly detection as a "noise-to-norm" paradigm, in which anomalies are identified as inconsistencies between a query image and its flawless approximation. Our pipeline achieves this by restoring the anomalous regions from the noisy corrupted query image while keeping the normal regions unchanged. DiffusionAD includes a denoising sub-network and a segmentation sub-network, which work together to provide intuitive anomaly detection and localization in an end-to-end manner, without the need for complicated post-processing steps. Remarkably, during inference, this framework delivers satisfactory performance with just one diffusion reverse process step, which is tens to hundreds of times faster than general diffusion methods. Extensive evaluations on standard and challenging benchmarks including VisA and DAGM show that DiffusionAD outperforms current state-of-the-art paradigms, demonstrating the effectiveness and generalizability of the proposed pipeline.

show abstract

Section: Related Workmentioning

confidence: 99%

DiffusionAD: Denoising Diffusion for Anomaly Detection

Zhang¹,

Wang²,

Wu³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Based on this mechanism, [3] introduces the distance measurement into the loss function and the measure of the outliers. On the basis of Autoencoder, [8], [34] and [39] restore images whose information have been erased due to the preprocessing (including rotation, removal of channels, cutout [7], etc.). CSI [33] introduces contrastive learning into anomaly detection, uses distributedally-shifted image as a negative sample to alleviate the performance penalty problem mentioned in [5].…”

Section: Related Workmentioning

confidence: 99%

DA-LSTM-VAE: Dual-Stage Attention-Based LSTM-VAE for KPI Anomaly Detection

Zhao

Zhang

Shang

et al. 2022

Entropy

View full text Add to dashboard Cite

To ensure the normal operation of the system, the enterprise’s operations engineer will monitor the system through the KPI (key performance indicator). For example, web page visits, server memory utilization, etc. KPI anomaly detection is a core technology, which is of great significance for rapid fault detection and repair. This paper proposes a novel dual-stage attention-based LSTM-VAE (DA-LSTM-VAE) model for KPI anomaly detection. Firstly, in order to capture time correlation in KPI data, long–short-term memory (LSTM) units are used to replace traditional neurons in the variational autoencoder (VAE). Then, in order to improve the effect of KPI anomaly detection, an attention mechanism is introduced into the input stage of the encoder and decoder, respectively. During the input stage of the encoder, a time attention mechanism is adopted to assign different weights to different time points, which can adaptively select important input sequences to avoid the influence of noise in the data. During the input stage of the decoder, a feature attention mechanism is adopted to adaptively select important latent variable representations, which can capture the long-term dependence of time series better. In addition, this paper proposes an adaptive threshold method based on anomaly scores measured by reconstruction probability, which can minimize false positives and false negatives and avoid adjustment of the threshold manually. Experimental results in a public dataset show that the proposed method in this paper outperforms other baseline methods.

show abstract

“…They have a powerful ability to extract high-level features. DNNs [9][10][11][12][13][14][15] have also gained great improvements in the task of defect detection compared with traditional methods. Most existing DNN-based inspection methods are based on supervised learning, which implies that a large number of manually annotated data are required.…”

Section: Introductionmentioning

confidence: 99%

Neural Subspace Learning for Surface Defect Detection

Liu

Chen

et al. 2022

Mathematics

View full text Add to dashboard Cite

Surface defect inspection is a key technique in industrial product assessments. Compared with other visual applications, industrial defect inspection suffers from a small sample problem and a lack of labeled data. Therefore, conventional deep-learning methods depending on huge supervised samples cannot be directly generalized to this task. To deal with the lack of labeled data, unsupervised subspace learning provides more clues for the task of defect inspection. However, conventional subspace learning methods focus on studying the linear subspace structure. In order to explore the nonlinear manifold structure, a novel neural subspace learning algorithm is proposed by substituting linear operators with nonlinear neural networks. The low-rank property of the latent space is approximated by limiting the dimensions of the encoded feature, and the sparse coding property is simulated by quantized autoencoding. To overcome the small sample problem, a novel data augmentation strategy called thin-plate-spline deformation is proposed. Compared with the rigid transformation methods used in previous literature, our strategy could generate more reliable training samples. Experiments on real-world datasets demonstrate that our method achieves state-of-the-art performance compared with unsupervised methods. More importantly, the proposed method is competitive and has a better generalization capability compared with supervised methods based on deep learning techniques.

show abstract

Split-Brain Autoencoder Approach for Surface Defect Detection

Cited by 5 publications

References 12 publications

DiffusionAD: Denoising Diffusion for Anomaly Detection

DiffusionAD: Denoising Diffusion for Anomaly Detection

DA-LSTM-VAE: Dual-Stage Attention-Based LSTM-VAE for KPI Anomaly Detection

Neural Subspace Learning for Surface Defect Detection

Contact Info

Product

Resources

About