Sea-Surface Floating Small Target Detection by One-Class Classifier in Time-Frequency Feature Space

Shi, Sai-Nan; Shui, Peng‐Lang

doi:10.1109/tgrs.2018.2838260

Cited by 103 publications

(111 citation statements)

References 45 publications

Supporting

Mentioning

110

Contrasting

Order By: Relevance

“…The single feature detector, such as the normalized Hurst exponent (NHE) [10], usually cannot obtain a satisfactory detection result because the returns' information is not well utilized. It is a feasible approach to employ multiple features to improve detection performance [16]. In radar target detection, the separation capacity differences of each feature between the target and sea clutter are very difficult to evaluate, because this depends on many factors, such as the target type and sea clutter condition, etc.…”

Section: Feature Extraction and The Multi-feature Detector Basedmentioning

confidence: 99%

“…Therefore, it is an effective way to include the detection features by time-frequency analysis of the received radar returns and to extract the effective features from the time-frequency images that are conducive to the detection of floating small targets. Here, we only give a brief overview of the time-frequency features that have been derived in detail in [16] and apply these time-frequency features to the subsequent multi-feature detectors.…”

Section: A the Description Of Multiple Featuresmentioning

confidence: 99%

“…In general, the detection performance of the iForest-based detector is better and the detection stability is higher especially for floating small target detection. In the second experiment, the iForest-based detector is compared with the normalized Hurst exponent (NHE) detector [10], the tri-feature-based detector [14] and the timefrequency features detector [16]. For the observation time 0.512 seconds and 1.024 seconds, Table II has listed the four detectors' average probability of detection on twenty data sets when PFA is 0.001.…”

Section: A Comparison Between Iforest-based Detector and Other Anomamentioning

confidence: 99%

“…Via the long autoregressive model, the parameters of Doppler power spectra (DPS) of time series are mapped into Riemannian manifold [13], and then the detection is accomplished by Riemannian geometry method. Tri-feature-based detectors, including amplitude and Doppler three features [14], polarization information three features [15] and three time-frequency (TF) features [16], achieve fairly good detection performance for floating small targets detection. However, the fast convexhull learning algorithm can only use three detection features at most, because PFA of convexhull is hard to control precisely in high-dimensional space (feature dimension is larger than 3).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Sea-Surface Floating Small Target Detection by Multifeature Detector Based on Isolation Forest

Zhu

Jiang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

In this paper, a multi-feature detector based on isolation forest (iForest) algorithm is developed to detect floating small targets in sea clutter. The conventional multi-feature detector can only process three features or less. The proposed detector aims to break the limitation of feature dimensions' number of the existed feature-based detectors and to improve the detection performance. It transforms the detection of floating small target into an anomaly detection problem in a high-dimensional feature space, breaking the limitation of the number of features. Firstly, a modified isolation forest is constructed from multiple features extracted from sea clutter. Meanwhile, the relative Doppler coefficient of variation (RCV) is proposed and added into the feature library. Then, taking the average path length as detection statistic, the detection threshold is obtained by Monte-Carlo technique at the given false alarm probability (PFA). Finally, the final decision is made by comparing the path length calculated from the cell under test (CUT) of radar returns with the detection threshold. Detection performances are evaluated based on twenty measured IPIX radar data sets. The experiment results show that the multi-feature detector based on isolation forest can obtain a significant performance improvement and has lower computation cost compared with the existed detectors.

show abstract

Section: Feature Extraction and The Multi-feature Detector Basedmentioning

confidence: 99%

Section: A the Description Of Multiple Featuresmentioning

confidence: 99%

Section: A Comparison Between Iforest-based Detector and Other Anomamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Sea-Surface Floating Small Target Detection by Multifeature Detector Based on Isolation Forest

Zhu

Jiang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Panagopoulos et al [28] applied three distinct signal processing techniques, like signal averaging, morphological filtering and time-frequency analysis to detect target in sea clutter. Shi et al [29] made an effort by using smoothed pseudo-Wigner-Ville distribution (SPWVD) algorithm to enhance time frequency features of the given signal. Authors applied SPWVD to extract time series information at the Cell-Under-Test (CUT) as well as reference cells near the CUT that enabled estimation of the differences between target returns and the TF pattern of sea clutter which was congregated on the normalized SPWVD.…”

Section: Related Workmentioning

confidence: 99%

Hanning Weighted Window Analysis Assisted Time-Series Analysis Model for Slow Moving Target Detection in Sea Clutter

Rajesh¹,

Udayarani²,

Jayaramaiah³

2020

International Journal of Engineering Research and Technology

View full text Add to dashboard Cite

The exponential rise in maritime movement, security threats from up-surging terrorism and smuggling has alarmed academia-industries to develop more efficient method for target detection in coastal area. This task becomes more complex and intricate in case of small moving target in seaclutter. Amongst the major available approaches such as multiple-radar based methods, wavelet analysis methods, Doppler measurements, Fourier transforms, Short Term Fourier Transform (STFT) based techniques, STFT has been found more robust. The ability to perform simultaneous time and frequency analysis for time-series assessment enables STFT to be used in small target detection in sea clutter; however the inherent limitations such as improper Windowing, window-size, number of samples, overlapping conditions affect overall performance. Considering it as motivation, in this paper we have developed an efficient Hanning Weighted Window Function (HWWF) model to be used in conjunction with STFT to perform Hanning-Weighted Overlapped Time-Series Analysis (HWOTSA) to detect slow moving target detection in sea clutter. The use of orthonormal transformation also called rotation enabled Hanning Window Analysis to use suitable STFT parameters and statistical test across the windows to achieve better accuracy of the target detection. HWOTSA assisted STFT in conjunction with the Probability Distribution Function Projection (PPM) over extracted features enabled accurate slow moving target detection in sea clutter. The MATLAB based simulation affirmed that the proposed method enables accurate and swift (slow) moving target detection in sea clutter, without introducing higher computational overheads and complexities. The estimation of Doppler measurement based target velocity estimation armour proposed method to be used in real-time applications for swift decision making.

show abstract

Floating Small Target Detection in Sea Clutter Based on Jointed Features in FRFT Domain

Shi

Zhang

Liu

2019

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Sea-Surface Floating Small Target Detection by One-Class Classifier in Time-Frequency Feature Space

Cited by 103 publications

References 45 publications

Sea-Surface Floating Small Target Detection by Multifeature Detector Based on Isolation Forest

Sea-Surface Floating Small Target Detection by Multifeature Detector Based on Isolation Forest

Hanning Weighted Window Analysis Assisted Time-Series Analysis Model for Slow Moving Target Detection in Sea Clutter

Floating Small Target Detection in Sea Clutter Based on Jointed Features in FRFT Domain

Contact Info

Product

Resources

About