ViV-Ano: Anomaly Detection and Localization Combining Vision Transformer and Variational Autoencoder in the Manufacturing Process

Abstract

The goal of image anomaly detection is to determine whether there is an abnormality in an image. Image anomaly detection is currently used in various fields such as medicine, intelligent information, military fields, and manufacturing. The encoder–decoder structure, which learns a normal-looking periodic normal pattern and shows good performance in judging anomaly scores through reconstruction errors showing the differences between the reconstructed images and the input image, is widely used in the field of anomaly detection. The existing image anomaly detection method extracts normal information through local features of the image, but the vision transformer base and the probability distribution are generated by learning the global relationship between image anomaly detection and an image patch that can locate anomalies to extract normal information. We propose Vision Transformer and VAE for Anomaly Detection (ViV-Ano), an anomaly detection model that combines a model variational autoencoder (VAE) with Vision Transformer (ViT). The proposed ViV-Ano model showed similar or better performance when compared to the existing model on a benchmark dataset. In addition, an MVTec anomaly detection dataset (MVTecAD), a dataset for industrial anomaly detection, showed similar or improved performance when compared to the existing model.

Enhanced Anomaly Detection in Manufacturing Processes Through Hybrid Deep Learning Techniques

“…That is, it trains on normal data only and detects abnormal patterns in parts that cannot be reconstructed [13]. The input data (i.e., process sensor data) require sequential information flow characteristics [39]. Therefore, we develop an LSTM-autoencoder that combines an autoencoder capable of detecting anomalies with LSTM capable of reflecting sequential time-series information.…”

“…4). This approach is suitable for environments that receive real-time data from manufacturing processes and detect anomalies [39,40].…”

Concept of the Intelligent Support of Decision Making for Manufacturing a 3D-Printed Hand Exoskeleton within Industry 4.0 and Industry 5.0 Paradigms

“…Development of clinical guidelines, indications, and contraindications for the use of the hand exoskeleton according to the paradigm of evidence-based medicine, including a combined approach [39] and possible secondary modifications [40][41][42]; • Development and marketing of a commercial hand exoskeleton (with support for all five fingers) following randomized clinical trials on a large group of patients; • Lightweight and environmentally friendly material and manufacturing technologies for personalized medical devices in compliance with the Medical Device Regulations (MDR) and ISO 13485 (quality control and life cycle monitoring of medical products from conception to complete withdrawal from the market); • Advanced artificial intelligence solutions for data analysis and control [43][44][45], including brain-computer interfaces [46][47][48];…”