Transfer Learning in Automated Gamma Spectral Identification

TL;DR

This paper explores transfer learning by applying CNN models trained on modeled gamma spectral data to classify real measured spectra, aiming to improve automated isotope identification across different data domains.

Contribution

It demonstrates the feasibility of transferring knowledge from modeled to measured spectral data using CNNs, enhancing spectral classification in practical scenarios.

Findings

Transfer learning enables CNNs trained on modeled data to classify measured spectra effectively.

The approach reduces the need for extensive measured data for training.

Results show promising accuracy in cross-domain spectral classification.

Abstract

The models and weights of prior trained Convolutional Neural Networks (CNN) created to perform automated isotopic classification of time-sequenced gamma-ray spectra, were utilized to provide source domain knowledge as training on new domains of potential interest. The previous results were achieved solely using modeled spectral data. In this work we attempt to transfer the knowledge gained to the new, if similar, domain of solely measured data. The ability to train on modeled data and predict on measured data will be crucial in any successful data-driven approach to this problem space.