Robust CFAR Ship Detector Based on Bilateral-Trimmed-Statistics of Complex Ocean Scenes in SAR Imagery: A Closed-Form Solution

Ai, Jiaqiu; Mao, Yuxiang; Luo, Qiwu; Xing, Mengdao; Jiang, Keyong; Jia, Lu; Yang, Xiaoxia

doi:10.1109/taes.2021.3050654

Cited by 57 publications

(25 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regardless of the single-polarized SAR image or the polarimetric SAR image, the target detection performance depends on the information used to discriminate targets from the background clutter. For the single-polarized SAR images, the most widely used detection feature is the amplitude information, based on which the constant false alarm rate (CFAR) detectors [1][2][3][4][5] were well studied in literatures. In recent years, the deep learning techniques [6][7][8][9][10] have been applied to SAR amplitude image target detection, such as the convolutional neural networks (CNN) [11], the faster regions with CNN (Faster-RCNN) [12], single shot multibox detectors (SSD) [13], and you only look once (YOLO) models [14,15].…”

Section: Introductionmentioning

confidence: 99%

“…For example, the deep detection networks may suffer performance decay dealing with strong interferences, large-scale imaging scenes, and multi-target situations with various targets of different sizes, especially for the detection of small-sized targets. In practice, the CFAR detectors [3][4][5] are still widely used due to its simplicity and adaptivity.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Outlier-Robust Superpixel-Level CFAR Detector With Truncated Clutter for Single Look Complex SAR Images

Peng

Shi

2022

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

View full text Add to dashboard Cite

The constant false alarm rate (CFAR) detectors are well studied for ship detection in SAR images, which suffer performance degradation due to the capture effect from interfering outliers, such as nearby targets, sidelobes and ghosts in multi-target environments. To address this issue, the clutter truncation scheme is adopted to reduce the outlier contamination in clutter samples such that the accuracy of clutter modeling can be improved. However, the selection of clutter truncation depth is difficult, which often resorts to sensitivity study. In this paper, the complex signal kurtosis (CSK) is first utilized as a statistical indicator for the decision of truncation depth to guarantee that the true clutter samples are maintained. Besides, a coarse-tofine detection process is designed, including global superpixel proposal with the CSK and local identification of target pixels with the superpixel-level CFAR detector based on truncated statistics. During the local CFAR detection stage, the segmented superpixels provide convenient sample indexing for the iterative clutter truncation processing. The elevated performance achieved by the proposed method mainly benefits from the schemes of two-stage detection and automatic clutter truncation, yielding the increased detection efficiency and accuracy at the same time. Besides, false alarms caused by radio frequency interference can be reduced. In the experiment, the comparative results with stateof-the-art methods based on the Sentinel-1 and Gaofen-3 SAR data validate the performance of the proposed method.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Outlier-Robust Superpixel-Level CFAR Detector With Truncated Clutter for Single Look Complex SAR Images

Peng

Shi

2022

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

View full text Add to dashboard Cite

show abstract

“…Outlier robust CFAR [13] uses the adaptively threshold to effectively removed the highintensity outliers and maintained real clutter samples. Bilateral-trimmed-statistics-based CFAR [14] uses the depth adaptive bilateral threshold to remove the both high and low intensity outliers, and greatly maintain the real clutter samples.…”

Section: Introductionmentioning

confidence: 99%

A False Alarm Elimination Algorithm of Foreign Objects Debris Detection Based on Duffing Oscillator

et al. 2022

View full text Add to dashboard Cite

The main challenge of foreign objects debris (FOD) monitoring based on millimeter-wave (MMW) radar is the interference of strong ground clutter. The clutter-map constant false alarm rate (CM-CFAR) method is commonly utilized for FOD monitoring, but detection results are usually accompanied by many false alarms in the complex ground clutter environment. To solve this problem, this work proposed a false alarm elimination algorithm. Specifically, the CM-CFAR method is firstly used to separate the suspicious targets (including FOD and false alarms) from the background clutter. Then, the Duffing detection system is constructed to detect the sinusoidal frequencies associated with suspicious targets. Finally, according to the detection results of Duffing system, suspicious targets can be classified as FOD and false alarms. In addition, in order to detect the sinusoidal frequencies efficiently, a quantitative detection method based on Poincare section points analysis is proposed to identify the state transition of Duffing oscillator. The results from both simulation and field case show the excellent false alarm elimination performance of the proposed algorithm.INDEX TERMS Foreign objects debris (FOD), millimeter-wave (MMW) radar, false alarm elimination, Duffing oscillator, quantitative detection method.

show abstract

“…To improve the safety of low-flying objects equipped with millimeter-wave radar, a first-order difference (FOD)-CFAR method is proposed in [41]. Furthermore, the CFAR detectors have wide applications in Synthetic aperture radar (SAR) imagery [42]- [44].…”

Section: Introductionmentioning

confidence: 99%

Coherent Radar Target Detection With In-Band Cyclostationary Wireless Interference

et al. 2022

View full text Add to dashboard Cite

Spectral congestion necessitates the in-band operation or the spectrum-sharing of legacy radar and communication systems. Since these systems operate in the same band in spectrum-sharing mode, they interfere with one another. To address this problem from the radar's perspective, this paper considers the coherent detection of target-reflected radar signals in the presence of interference from an in-band cyclostationary digital modulated wireless communication signal. Three different cases of targetreflected radar signals, namely, deterministic signals, signals with random phase, and completely random signals, are considered in this paper. The optimum detection rules are derived for these three cases and the corresponding receiver structures for the equalization of the interfering signal are presented. Suboptimum detection structures are also derived with the assumption that the in-band interference is a white stationary time-invariant Gaussian process. Further, considering the equalization, modified CFAR receiver structures are also presented. By considering the mathematical models for cyclostationary or periodic inband interference, the performances of the optimum, sub-optimum detectors, and modified CFAR detectors are quantified analytically in terms of detection probability and false alarm probability, and the resulting receiver operating characteristic (ROC) curves are analyzed as a function of the signal-to-interference ratio. It is demonstrated that improper equalization of the interfering signal significantly affects the performance of the optimum detector and this impact is analyzed in detail. As spectrum-sharing becomes more prevalent due to spectrum congestion, the proposed optimal, sub-optimal, and modified CFAR detection rules and receiver structures can be incorporated into existing systems with substantial savings.

show abstract

Robust CFAR Ship Detector Based on Bilateral-Trimmed-Statistics of Complex Ocean Scenes in SAR Imagery: A Closed-Form Solution

Cited by 57 publications

References 45 publications

Outlier-Robust Superpixel-Level CFAR Detector With Truncated Clutter for Single Look Complex SAR Images

Outlier-Robust Superpixel-Level CFAR Detector With Truncated Clutter for Single Look Complex SAR Images

A False Alarm Elimination Algorithm of Foreign Objects Debris Detection Based on Duffing Oscillator

Coherent Radar Target Detection With In-Band Cyclostationary Wireless Interference

Contact Info

Product

Resources

About