A Dual-Branch Ensemble Learning Method for Industrial Anomaly Detection: Fusion and Optimization of Scattering and PCA Features

Industrial visual anomaly detection remains challenging because practical inspection systems must achieve high detection accuracy while operating under highly imbalanced data, diverse defect patterns, limited computational resources, and increasing demands for interpretability. This work aims to develop a lightweight yet effective and explainable anomaly detection framework for industrial images in settings where a limited number of labeled anomalous samples are available. We propose a dual-branch feature-based supervised ensemble method that integrates complementary representations: a PCA branch to capture linear global structure and a scattering branch to model multi-scale textures. A heterogeneous pool of classical learners (SVM, RF, ET, XGBoost, and LightGBM) is trained on each feature branch, and stable probability outputs are obtained via stratified K-fold out-of-fold training, probability calibration, and a quantile-based threshold search. Decision-level fusion is then performed by stacking, where logistic regression, XGBoost, and LightGBM serve as meta-learners over the out-of-fold probabilities of the selected top-K base learners. Experiments on two public benchmarks (MVTec AD and BTAD) show that the proposed method substantially improves the best PCA-based single model, achieving relative F1_score gains of approximately 31% (MVTec AD) and 26% (BTAD), with maximum AUC values of about 0.91 and 0.96, respectively, under comparable inference complexity. Overall, the results demonstrate that combining high-quality handcrafted features with supervised ensemble fusion provides a practical and interpretable alternative/complement to heavier deep models for resource-constrained industrial anomaly detection, and future work will explore more category-adaptive decision strategies to further enhance robustness on challenging classes.

• Publication year

  2026

• Venue

  Applied Sciences

• Publication date

  2026-02-05

• Fields of study

  Not labeled

• Identifiers
  DOI 10.3390/app16031597
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Robust Industrial Surface Defect Detection Using Statistical Feature Extraction and Capsule Network Architectures

  2025cited by this paper
• Enhancing anomaly detection in IoT-driven factories using Logistic Boosting, Random Forest, and SVM: A comparative machine learning approach

  2025cited by this paper
• Enhancing ML-based anomaly detection in data management for security through integration of IoT, cloud, and edge computing

  2025cited by this paper
• A survey on RGB, 3D, and multimodal approaches for unsupervised industrial image anomaly detection

  2025cited by this paper
• Selective ensemble method for anomaly detection based on parallel learning

  2024cited by this paper
• Using Ensemble Learning for Anomaly Detection in Cyber–Physical Systems

  2024cited by this paper
• Graph Regularization Methods in Soft Detector Fusion

  2023cited by this paper
• A Unified Theory of Diversity in Ensemble Learning

  2023cited by this paper
• Unsupervised Machine Learning Techniques for Detecting PLC Process Control Anomalies

  2022cited by this paper
• Evolutionary bagging for ensemble learning

  2022cited by this paper
• Anomaly Detection for a Water Treatment System Based on One-Class Neural Network

  2022cited by this paper
• An Improved KNN-Based Efficient Log Anomaly Detection Method with Automatically Labeled Samples

  2021cited by this paper
• Machine Learning for Anomaly Detection: A Systematic Review

  2021cited by this paper
• Generalized isolation forest for anomaly detection

  2021cited by this paper
• A Comprehensive Survey of Anomaly Detection Algorithms

  2021cited by this paper
• Deep Learning in the Industrial Internet of Things: Potentials, Challenges, and Emerging Applications

  2021cited by this paper
• The MVTec Anomaly Detection Dataset: A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection

  2021cited by this paper
• Deep Learning for Anomaly Detection

  2020cited by this paper
• Improving security using SVM-based anomaly detection: issues and challenges

  2020cited by this paper
• Predictive modelling for solar thermal energy systems: A comparison of support vector regression, random forest, extra trees and regression trees

  2018cited by this paper
• Anomaly detection model based on data stream clustering

  2017cited by this paper
• A survey of intrusion detection systems based on ensemble and hybrid classifiers

  2017cited by this paper
• Optimal ROC-Based Classification and Performance Analysis under Bayesian Uncertainty Models

  2016cited by this paper
• Hybrid and Ensemble Methods in Machine Learning

  2013cited by this paper
• Invariant Scattering Convolution Networks

  2013cited by this paper
• A Review on Ensembles for the Class Imbalance Problem: Bagging-, Boosting-, and Hybrid-Based Approaches

  2012cited by this paper
• Group Invariant Scattering

  2011cited by this paper
• Principal component analysis

  2010cited by this paper
• Plastic card fraud detection using peer group analysis

  2008cited by this paper
• Ten Years of Industrie 4.0

  year unknowncited by this paper

• No citing papers are available for this paper.