AC-LSTM: Anomaly State Perception of Infrared Point Targets Based on CNN+LSTM
Open Access
- 4 July 2022
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 14 (13) , 3221
Abstract
Anomaly perception of infrared point targets has high application value in many fields, such as maritime surveillance, airspace surveillance, and early warning systems. This kind of abnormality includes the explosion of the target, the separation between stages, the disintegration caused by the abnormal strike, etc. By extracting the radiation characteristics of continuous frame targets, it is possible to analyze and warn the target state in time. Most anomaly detection methods adopt traditional outlier detection, which has the problems of poor accuracy and a high false alarm rate. Driven by data, this paper proposes a new network structure, called AC-LSTM, which combines Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM), and embeds the Periodic Time Series Data Attention module (PTSA). The network can better extract the spatial and temporal characteristics of one-dimensional time series data, and the PTSA module can consider the periodic characteristics of the target in the process of continuous movement, and focus on abnormal data. In addition, this paper also proposes a new time series data enhancement method, which slices and re-amplifies the long time series data. This method significantly improves the accuracy of anomaly detection. Through a large number of experiments, AC-LSTM has achieved higher scores on our collected datasets than other methods.Keywords
Funding Information
- Science and Technology Department of Jilin Province, China (20210201137GX)
This publication has 31 references indexed in Scilit:
- CNN-LSTM Neural Network Model for Quantitative Strategy Analysis in Stock Markets2017
- Convolutional neural networks for time series classificationJournal of Systems Engineering and Electronics, 2017
- Collective Anomaly Detection Based on Long Short-Term Memory Recurrent Neural Networks2016
- Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks2015
- A Review of Machine Learning based Anomaly Detection TechniquesInternational Journal of Computer Applications Technology and Research, 2013
- Isolation-Based Anomaly DetectionACM Transactions on Knowledge Discovery From Data, 2012
- Anomaly detectionACM Computing Surveys, 2009
- Online Anomaly Prediction for Robust Cluster SystemsInternational Conference on Data Engineering, 2009
- Hidden semi-Markov model for anomaly detectionApplied Mathematics and Computation, 2008
- Long Short-Term MemoryNeural Computation, 1997