Robust Classification Scheme for Airplane Targets With Low Resolution Radar Based on EMD-CLEAN Feature Extraction Method

Du, Lan; Wang, Baoshuai; Li, Yanbing; Liu, Hongwei

doi:10.1109/jsen.2013.2272119

Cited by 48 publications

(19 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[14] proposes a hierarchical micro-Doppler feature extraction method to categorize moving vehicles into two kinds, i.e., wheeled vehicle and tracked vehicle. [15][16][17][18] analyze the JEM characteristics of the airplane targets and investigate the airplane classification problem based on the JEM effects, of which [15,17] analyze the scattering characteristics of the rotating blades in a helicopter, [16] utilizes the measured JEM characteristics to realize the radar target identification and [18] proposes a new feature extraction method based on Empirical Mode Decomposition (EMD) and CLEAN technique. Compared with the features used with high resolution radar [19][20][21][22], the micro-Doppler features represent the different micro-motion characteristics from different target categories for target classification (which means to decide the target category), rather than describe the detailed waveforms of the training echoes from some given target models for target recognition (which means to decide the detailed target model), to guarantee the generalization performance among different target models from the same target category.…”

Section: Introductionmentioning

confidence: 99%

Micro-Doppler Feature Extraction Based on Time-Frequency Spectrogram for Ground Moving Targets Classification With Low-Resolution Radar

Wang

et al. 2016

IEEE Sensors J.

Self Cite

View full text Add to dashboard Cite

A novel feature extraction method based on micro-Doppler signature is proposed to categorize ground moving targets into three kinds, i.e., single walking person, two people walking and a moving wheeled vehicle. Signal models and measured data from a low resolution radar are first analyzed to find the differences between the micro-Doppler signatures from the three kinds of considered targets. Then such discriminative micro-Doppler signatures are represented via a 3-dimensional feature vector extracted from the time-frequency spectrograms.In the experiments based on the measured data, the ratio of the between-class distance to the within-class distance, which is defined based on Fisher Discriminant Analysis (FDA), is exploited to assess the discriminative ability of the 3-dimensional feature vector. Moreover, Support Vector Machine (SVM) classifier is utilized to evaluate the classification performance. Experimental results show that the proposed micro-Doppler features can achieve a good discriminative ability and a satisfactory classification performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Micro-Doppler Feature Extraction Based on Time-Frequency Spectrogram for Ground Moving Targets Classification With Low-Resolution Radar

Wang

et al. 2016

IEEE Sensors J.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The parameter of Gaussian kernel is tuned in the range [0, 5] for each classification method, and we select the best classification result from those of the different parameter settings for the following analysis and comparison. Figure 3 depicts the classification accuracies versus SNR, generated via the four classification schemes, including classification with the CPPCA-BIC based noise reduction method, classification with the CLEAN based noise reduction method and the accurate SNR information [5] , classification with the CLEAN based noise reduction method and the ±3dB biased SNR information, and classification without noise reduction. For simplicity, the above four classification schemes are referred to as CPPCA-BIC, CLEAN, CLEAN with Bias and Noised Features respectively in Figure 3.…”

Section: Resultsmentioning

confidence: 99%

“…As shown in the experimental results in [5], the classification performance dramatically deteriorates under the low SNR condition. Since the SNR directly relates to the distance between the target and radar for a given noise power and radar power, the noise robustness of a classification method is an important factor in increasing the classification distance between the target and radar in the real application.…”

Section: Introductionmentioning

confidence: 84%

“…Since the SNR directly relates to the distance between the target and radar for a given noise power and radar power, the noise robustness of a classification method is an important factor in increasing the classification distance between the target and radar in the real application. Based on the sparse representation theory, a simple noise reduction method is proposed in [5]. Nevertheless, this method requires the prior information about SNR.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A noise-robust radar target classification method based on complex probabilistic principal component analysis

et al. 2014

2014 XXXIth URSI General Assembly and Scientific Symposium (URSI GASS)

Self Cite

View full text Add to dashboard Cite

We develop a noise-robust radar target classification method to discriminate the moving vehicle and walking human. The traditional real-valued Probabilistic Principal Component Analysis (PPCA) model is extended to the complex-value domain for modeling the low resolution radar echoes from the ground moving targets. The denoising preprocessing is accomplished by signal reconstruction with the proposed Complex Probabilistic Principal Component Analysis (CPPCA) model, where we utilize the Bayesian Inference Criterion (BIC) to adaptively select the principal components. After denoising, a 3-dimensional timefrequency feature vector is extracted from the denoised micro-Doppler signatures of the two kinds of ground targets, and the classification is performed via Support Vector Machine (SVM) classifier. In the experiments based on the measured data, the proposed classification scheme shows the good classification and denoising performance under the relatively low SNR condition. In the real application, the advantage in SNR can effectively extend the classification distance between the target and radar.

show abstract

“…It should be pointed out that when extracting the micromotion feature of a target, we need to satisfy PRF > 8πfR max ∕λ to avoid the frequency-domain aliasing of the micro-Doppler signal, 16 where f represents the target rotation frequency, R max denotes the maximum rotation radius of the target, and λ is the wavelength. Obviously, the low PRF of tracking pulses usually cannot satisfy this requirement, which leads to under-sampling of the micro-Doppler signal and produces frequency-domain aliasing.…”

Section: 23mentioning

confidence: 99%

Micromotion feature extraction of radar target using tracking pulses with adaptive pulse repetition frequency adjustment

Chen

Zhang

et al. 2014

J. Appl. Remote Sens

View full text Add to dashboard Cite

Abstract. In multifunction phased array radar systems, different activities (e.g., tracking, searching, imaging, feature extraction, recognition, etc.) would need to be performed simultaneously. To relieve the conflict of the radar resource distribution, a micromotion feature extraction method using tracking pulses with adaptive pulse repetition frequencies (PRFs) is proposed in this paper. In this method, the idea of a varying PRF is utilized to solve the frequency-domain aliasing problem of the micro-Doppler signal. With appropriate atom set construction, the micromotion feature can be extracted and the image of the target can be obtained based on the Orthogonal Matching Pursuit algorithm. In our algorithm, the micromotion feature of a radar target is extracted from the tracking pulses and the quality of the constructed image is fed back into the radar system to adaptively adjust the PRF of the tracking pulses. Finally, simulation results illustrate the effectiveness of the proposed method. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

show abstract

Robust Classification Scheme for Airplane Targets With Low Resolution Radar Based on EMD-CLEAN Feature Extraction Method

Cited by 48 publications

References 40 publications

Micro-Doppler Feature Extraction Based on Time-Frequency Spectrogram for Ground Moving Targets Classification With Low-Resolution Radar

Micro-Doppler Feature Extraction Based on Time-Frequency Spectrogram for Ground Moving Targets Classification With Low-Resolution Radar

A noise-robust radar target classification method based on complex probabilistic principal component analysis

Micromotion feature extraction of radar target using tracking pulses with adaptive pulse repetition frequency adjustment

Contact Info

Product

Resources

About