scholarly journals A Novel Multi-Input Bidirectional LSTM and HMM Based Approach for Target Recognition from Multi-Domain Radar Range Profiles

Electronics ◽  
2019 ◽  
Vol 8 (5) ◽  
pp. 535 ◽  
Author(s):  
Fei Gao ◽  
Teng Huang ◽  
Jun Wang ◽  
Jinping Sun ◽  
Amir Hussain ◽  
...  
Keyword(s):  
Frequency Domain ◽  
Short Term Memory ◽  
Target Identification ◽  
Target Recognition ◽  
Recognition Performance ◽  
Recognition Task ◽  
Identification Accuracy ◽  
Time Frequency ◽  
Sequence Features ◽  
Novel Target

Radars, as active detection sensors, are known to play an important role in various intelligent devices. Target recognition based on high-resolution range profile (HRRP) is an important approach for radars to monitor interesting targets. Traditional recognition algorithms usually rely on a single feature, which makes it difficult to maintain the recognition performance. In this paper, 2-D sequence features from HRRP are extracted in various data domains such as time-frequency domain, time domain, and frequency domain. A novel target identification method is then proposed, by combining bidirectional Long Short-Term Memory (BLSTM) and a Hidden Markov Model (HMM), to learn these multi-domain sequence features. Specifically, we first extract multi-domain HRRP sequences. Next, a new multi-input BLSTM is proposed to learn these multi-domain HRRP sequences, which are then fed to a standard HMM classifier to learn multi-aspect features. Finally, the trained HMM is used to implement the recognition task. Extensive experiments are carried out on the publicly accessible, benchmark MSTAR database. Our proposed algorithm is shown to achieve an identification accuracy of over 91% with a lower false alarm rate and higher identification confidence, compared to several state-of-the-art techniques.

scholarly journals Micro-Doppler-Based Space Target Recognition with a One-Dimensional Parallel Network

2020 ◽  
Vol 2020 ◽  
pp. 1-10
Author(s):  
Lixun Han ◽  
Cunqian Feng
Keyword(s):  
Short Term Memory ◽  
Target Identification ◽  
Target Recognition ◽  
Parallel Structure ◽  
Recognition Method ◽  
One Dimensional ◽  
Time Frequency ◽  
Physical Attributes ◽  
Frequency Modulations ◽  
Space Target

Space target identification is key to missile defense. Micromotion, as an inherent attribute of the target, can be used as the theoretical basis for target recognition. Meanwhile, time-varying micro-Doppler (m-D) frequency shifts induce frequency modulations on the target echo, which can be referred to as the m-D effect. m-D features are widely used in space target recognition as it can reflect the physical attributes of the space targets. However, the traditional recognition method requires human participation, which often leads to misjudgment. In this paper, an intelligent recognition method for space target micromotion is proposed. First, accurate and suitable models of warhead and decoy are derived, and then the m-D formulae are offered. Moreover, we present a deep-learning (DL) model composed of a one-dimensional parallel structure and long short-term memory (LSTM). Then, we utilize this DL model to recognize time-frequency distribution (TFD) of different targets. Finally, simulations are performed to validate the effectiveness of the proposed method.

scholarly journals Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Jianghua Nie ◽  
Yongsheng Xiao ◽  
Lizhen Huang ◽  
Feng Lv
Keyword(s):  
Least Squares ◽  
Data Augmentation ◽  
Short Term Memory ◽  
Target Recognition ◽  
Signal To Noise Ratio ◽  
Recognition Performance ◽  
Feature Learning ◽  
Generative Adversarial Network ◽  
Time Frequency ◽  
Adversarial Network

Aiming at the problem of radar target recognition of High-Resolution Range Profile (HRRP) under low signal-to-noise ratio conditions, a recognition method based on the Constrained Naive Least-Squares Generative Adversarial Network (CN-LSGAN), Short-time Fourier Transform (STFT), and Convolutional Neural Network (CNN) is proposed. Combining the Least-Squares Generative Adversarial Network (LSGAN) with the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP), the CN-LSGAN is presented and applied to the HRRP denoise. The frequency domain and phase features of HRRP are gained by STFT in order to facilitate feature learning and also match the input data format of the CNN. These experimental results show that the CN-LSGAN has better data augmentation performance and can effectively avoid the model collapse compared to the generative adversarial network (GAN) and LSGAN. Also, the method has better recognition performance than the one-dimensional CNN method and the Long Short-Term Memory (LSTM) network method.

Transient Resonance Scattering and Target Identification

1997 ◽  
Vol 50 (3) ◽  
pp. 131-148 ◽  
Author(s):  
G. C. Gaunaurd ◽  
H. C. Strifors
Keyword(s):  
Frequency Domain ◽  
Target Identification ◽  
Resonance Scattering ◽  
Convergent Series ◽  
Processing Technique ◽  
Signal Processing Technique ◽  
Time Frequency ◽  
Transient Scattering ◽  
Transient Resonance ◽  
Active Classification

The article presents an overview of transient resonance scattering, emphasizing one of its most important applications—the active classification of sonar and radar targets. It discusses traditional, classical techniques such as the Watson-Sommerfeld method (WSM) to transform classical, and slowly convergent normal-mode series in the frequency domain, to rapidly convergent series in the domain of the complex generalization, λ, of the mode-order, n. In view of its analytical complexity and the advent of computers that can overcome slow convergence difficulties, the WSM is not as popular today as it once was. Its main advantage remains its ability to extract physical interpretations from the mathematical results. Resonance scattering focuses on the resonance spectral region of targets. Of these, the penetrable (ie, elastic or dielectric) ones are the subjects of main interest here, particularly those insonified/illuminated by (finite) pulses of various types. The authors describe the exact isolation and extraction of the resonances contained within the scattering cross-section of a penetrable target by subtraction of suitable, background, geometrical contributions. These backgrounds are often given by the solution for an identical, but impenetrable target. This seems to be the main usefulness of impenetrable target solutions in underwater acoustics, which, generally, are physically unrealistic idealizations. The resonances identify the target as its fingerprint. Examples are shown to illustrate various transient scattering phenomena in acoustics and electromagnetism. The article shows exactly how the broadband pulses emitted by an impulse sonar (or radar) extract a substantial number of resonances from the echoes of penetrable targets. Further, it is shown how these are actually used to identify all physical characteristics of various analyzed targets, thus, indeed identifying them. The application of a novel signal processing technique that analyzes the echoes in the joint time-frequency domain is examined. This shows much promise for target identification purposes. Many distributions of the Wigner-type were used by us to generate simulated and experimental echo-displays in time-frequency that show the advantages of the process. The present overview supplements two earlier ones [23, 48] on closely related subjects. The article includes 101 references.

scholarly journals Recognition of Emotions Using Multichannel EEG Data and DBN-GC-Based Ensemble Deep Learning Framework

2018 ◽  
Vol 2018 ◽  
pp. 1-11 ◽  
Author(s):  
Hao Chao ◽  
Huilai Zhi ◽  
Liang Dong ◽  
Yongli Liu
Keyword(s):  
Deep Learning ◽  
Frequency Domain ◽  
Time Domain ◽  
Recognition Task ◽  
Frequency Characteristics ◽  
Time Frequency ◽  
Learning Framework ◽  
Valence States ◽  
Correlation Information ◽  
Multiple Domains

Fusing multichannel neurophysiological signals to recognize human emotion states becomes increasingly attractive. The conventional methods ignore the complementarity between time domain characteristics, frequency domain characteristics, and time-frequency characteristics of electroencephalogram (EEG) signals and cannot fully capture the correlation information between different channels. In this paper, an integrated deep learning framework based on improved deep belief networks with glia chains (DBN-GCs) is proposed. In the framework, the member DBN-GCs are employed for extracting intermediate representations of EEG raw features from multiple domains separately, as well as mining interchannel correlation information by glia chains. Then, the higher level features describing time domain characteristics, frequency domain characteristics, and time-frequency characteristics are fused by a discriminative restricted Boltzmann machine (RBM) to implement emotion recognition task. Experiments conducted on the DEAP benchmarking dataset achieve averaged accuracy of 75.92% and 76.83% for arousal and valence states classification, respectively. The results show that the proposed framework outperforms most of the above deep classifiers. Thus, potential of the proposed framework is demonstrated.

Target identification in the time-frequency domain

1996 ◽  
Author(s):  
Ismail I. Jouny ◽  
Passant V. Karunaratne ◽  
Moeness G. Amin
Keyword(s):  
Frequency Domain ◽  
Target Identification ◽  
Time Frequency

scholarly journals An Application of Time-Frequency-Based Analysis Method for Identifying Bolt Anchorage System

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
H. Q. Zheng ◽  
Y. Zhang ◽  
G. Han ◽  
X. Y. Sun
Keyword(s):  
Frequency Domain ◽  
Time Domain ◽  
Identification Accuracy ◽  
Analysis Method ◽  
Defect Identification ◽  
Time Frequency ◽  
Reinforcing Bar ◽  
Anchorage System ◽  
Convolution Kernels ◽  
Domain Information

A rock bolt refers to a reinforcing bar used commonly in geotechnical engineering. Also, defect identification of bolt anchorage system determines the safe operation of the reinforced structures. In the present paper, to accurately extract defect information, a CNN model based on time-frequency analysis is proposed, covering both time-domain and frequency-domain information. The effect of the number of convolution kernels on the defect identification results is discussed. By laboratory experiments, the performances of STFT-based CNN with those of time-domain input or frequency-domain input-based 1D CNN are compared, and the results demonstrate that the proposed method showed enhanced performance in identification accuracy.

scholarly journals A Deep Learning-Based Satellite Target Recognition Method Using Radar Data

Sensors ◽  
2019 ◽  
Vol 19 (9) ◽  
pp. 2008
Author(s):  
Lu ◽  
Zhang ◽  
Xu ◽  
Lin ◽  
Huo
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Target Recognition ◽  
Recognition Performance ◽  
Recognition Task ◽  
Performance Testing ◽  
Radar Data ◽  
Data Partition ◽  
Distance Metric ◽  
Recognition Method

A novel satellite target recognition method based on radar data partition and deep learning techniques is proposed in this paper. For the radar satellite recognition task, orbital altitude is introduced as a distinct and accessible feature to divide radar data. On this basis, we design a new distance metric for HRRPs called normalized angular distance divided by correlation coefficient (NADDCC), and a hierarchical clustering method based on this distance metric is applied to segment the radar observation angular domain. Using the above technology, the radar data partition is completed and multiple HRRP data clusters are obtained. To further mine the essential features in HRRPs, a GRU-SVM model is designed and firstly applied for radar HRRP target recognition. It consists of a multi-layer GRU neural network as a deep feature extractor and linear SVM as a classifier. By training, GRU neural network successfully extracts effective and highly distinguishable features of HRRPs, and feature visualization technology shows its advantages. Furthermore, the performance testing and comparison experiments also demonstrate that GRU neural network possesses better comprehensive performance for HRRP target recognition than LSTM neural network and conventional RNN, and the recognition performance of our method is almost better than that of other several common feature extraction methods or no data partition.

scholarly journals Recognition of EEG Signal Motor Imagery Intention Based on Deep Multi-View Feature Learning

Sensors ◽  
2020 ◽  
Vol 20 (12) ◽  
pp. 3496
Author(s):  
Jiacan Xu ◽  
Hao Zheng ◽  
Jianhui Wang ◽  
Donglin Li ◽  
Xiaoke Fang
Keyword(s):  
Frequency Domain ◽  
Motor Imagery ◽  
Recognition Performance ◽  
Feature Learning ◽  
Feature Representation ◽  
Support Vector ◽  
Imagery Task ◽  
Learning Method ◽  
Eeg Signals ◽  
Time Frequency

Recognition of motor imagery intention is one of the hot current research focuses of brain-computer interface (BCI) studies. It can help patients with physical dyskinesia to convey their movement intentions. In recent years, breakthroughs have been made in the research on recognition of motor imagery task using deep learning, but if the important features related to motor imagery are ignored, it may lead to a decline in the recognition performance of the algorithm. This paper proposes a new deep multi-view feature learning method for the classification task of motor imagery electroencephalogram (EEG) signals. In order to obtain more representative motor imagery features in EEG signals, we introduced a multi-view feature representation based on the characteristics of EEG signals and the differences between different features. Different feature extraction methods were used to respectively extract the time domain, frequency domain, time-frequency domain and spatial features of EEG signals, so as to made them cooperate and complement. Then, the deep restricted Boltzmann machine (RBM) network improved by t-distributed stochastic neighbor embedding(t-SNE) was adopted to learn the multi-view features of EEG signals, so that the algorithm removed the feature redundancy while took into account the global characteristics in the multi-view feature sequence, reduced the dimension of the multi-visual features and enhanced the recognizability of the features. Finally, support vector machine (SVM) was chosen to classify deep multi-view features. Applying our proposed method to the BCI competition IV 2a dataset we obtained excellent classification results. The results show that the deep multi-view feature learning method further improved the classification accuracy of motor imagery tasks.

scholarly journals Music Composition and Emotion Recognition Using Big Data Technology and Neural Network Algorithm

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Yu Wang
Keyword(s):  
Neural Network ◽  
Big Data ◽  
Emotion Recognition ◽  
Frequency Domain ◽  
Music Composition ◽  
Feature Recognition ◽  
Short Term Memory ◽  
Learning Ability ◽  
Time Frequency ◽  
Big Data Technology

To implement a mature music composition model for Chinese users, this paper analyzes the music composition and emotion recognition of composition content through big data technology and Neural Network (NN) algorithm. First, through a brief analysis of the current music composition style, a new Music Composition Neural Network (MCNN) structure is proposed, which adjusts the probability distribution of the Long Short-Term Memory (LSTM) generation network by constructing a reasonable Reward function. Meanwhile, the rules of music theory are used to restrict the generation of music style and realize the intelligent generation of specific style music. Afterward, the generated music composition signal is analyzed from the time-frequency domain, frequency domain, nonlinearity, and time domain. Finally, the emotion feature recognition and extraction of music composition content are realized. Experiments show that: when the iteration times of the function increase, the number of weight parameter adjustments and learning ability will increase, and thus the accuracy of the model for music composition can be greatly improved. Meanwhile, when the iteration times increases, the loss function will decrease slowly. Moreover, the music composition generated through the proposed model includes the following four aspects: sadness, joy, loneliness, and relaxation. The research results can promote music composition intellectualization and impacts traditional music composition mode.

Sign in / Sign up

Export Citation Format

Share Document