Convolutional Ensembling based Few-Shot Defect Detection Technique

TL;DR

This paper introduces a convolutional ensembling approach for few-shot defect detection that leverages multiple pre-trained models to improve accuracy and efficiency, enabling real-time anomaly detection.

Contribution

It proposes a novel ensembling technique combining pre-trained convolutional models for few-shot classification, reducing parameters and boosting accuracy for defect detection.

Findings

Achieved 92.30% accuracy on a power-line defect dataset.

Outperformed existing state-of-the-art methods without additional tuning.

Demonstrated suitability for real-time defect detection applications.

Abstract

Over the past few years, there has been a significant improvement in the domain of few-shot learning. This learning paradigm has shown promising results for the challenging problem of anomaly detection, where the general task is to deal with heavy class imbalance. Our paper presents a new approach to few-shot classification, where we employ the knowledge-base of multiple pre-trained convolutional models that act as the backbone for our proposed few-shot framework. Our framework uses a novel ensembling technique for boosting the accuracy while drastically decreasing the total parameter count, thus paving the way for real-time implementation. We perform an extensive hyperparameter search using a power-line defect detection dataset and obtain an accuracy of 92.30% for the 5-way 5-shot task. Without further tuning, we evaluate our model on competing standards with the existing state-of-the-art methods and outperform them.