Scholarlink

<p>Anomaly detection is critical for identifying various malfunctions or irregularities in image data. We present a highly efficient approach in terms of memory space, energy consumption and runtime using a clustering method to improve the performance of anomaly detection models by identifying anomalous and non-anomalous image features. Our model is based on pre-trained deep neural network feature maps and a clustering algorithm that is trained on the MVTec anomaly detection dataset, a benchmark dataset for anomaly detection methods with a focus on industrial inspections. The results show that the model is also capable of dealing with images of the gastrointestinal Kvasir-Capsule dataset taken via wireless capsule endoscopy. The model’s binary classification results outperform comparable anomaly detection methods on the Kvasir-Capsule dataset obtained with deep learning with a precision of 81.46&#xa0;%, a recall of 76.01&#xa0;% and a F1&#xa0;score of 78.64&#xa0;%. At the same time our model is highly efficient by using only approximately 4 million parameters during inference.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Efficient anomaly detection with feature maps and clustering

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月8日

查看全文
  • Sophie Böttcher,
  • Gabriele Gühring

摘要

Anomaly detection is critical for identifying various malfunctions or irregularities in image data. We present a highly efficient approach in terms of memory space, energy consumption and runtime using a clustering method to improve the performance of anomaly detection models by identifying anomalous and non-anomalous image features. Our model is based on pre-trained deep neural network feature maps and a clustering algorithm that is trained on the MVTec anomaly detection dataset, a benchmark dataset for anomaly detection methods with a focus on industrial inspections. The results show that the model is also capable of dealing with images of the gastrointestinal Kvasir-Capsule dataset taken via wireless capsule endoscopy. The model’s binary classification results outperform comparable anomaly detection methods on the Kvasir-Capsule dataset obtained with deep learning with a precision of 81.46 %, a recall of 76.01 % and a F1 score of 78.64 %. At the same time our model is highly efficient by using only approximately 4 million parameters during inference.