Research on Radioisotope Identification of Nuclear Pulse Sequence Based on CNN-BiLSTM

Aiming at the statistical limitations of isotope identification Methods by gamma spectroscopy, this paper studies a method of isotope recognition based on improved Hilbert-Huang Transform (HHT) combined with convolutional bidirectional long short-term memory neural network (CNN-BiLSTM), which does not require spectrum formation and directly analyzes nuclear pulse sequences. Results show that the model can achieve 96.98 per cent recognition accuracy. In addition, the model shows a good recognition effect in different pulse number and pulse sequence time, which further verifies the feasibility of isotope recognition based on nuclear pulse sequence, and provides a new idea and method for radioisotope recognition technology.