An Adaptively Truncated Clutter-Statistics-Based Two-Parameter CFAR Detector in SAR Imagery

Ai, Jiaqiu; Yang, Xiu; Song, Ji-Tao; Zeng, Dong; Jia, Lu; Zhou, Fang

doi:10.1109/joe.2017.2768198

Cited by 56 publications

(35 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mean and variance of the clutter are used for determining the threshold T in the two-parameter CFAR [8]. Assuming that the clutter background obeys Gaussian distribution, the probability density function p(x) is…”

Section: B Two-parameter Cfar Methodsmentioning

confidence: 99%

See 1 more Smart Citation

SAR Target CFAR Detection Via GPU Parallel Operation

Cui

Quan

Cao

et al. 2018

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

View full text Add to dashboard Cite

The constant false alarm rate with convolution and pooling (CP-CFAR) method, which can improve the detection efficiency via GPU parallel acceleration in the airborne synthetic aperture radar (SAR) images, is proposed in this paper. The constant false alarm rate (CFAR) method is one of the most widely used methods for target detection in airborne SAR images. However, since the CFAR is performed on a pixel-by-pixel basis, the time consumption will increase rapidly with the expansion of image scene. Even if the GPU is used for acceleration, the efficiency improvement is still limited, which cannot meet the real-time processing requirements. Therefore, the CP-CFAR method for the target detection of SAR images is proposed in this paper. The convolution layer uses the horizontal and vertical Sobel operators to improve the contrast between targets and background, and the pooling layer can reduce the processing dimension of the images. The convolution and pooling layers are added before the twoparameter CFAR, which can reduce the computational elements but without losing the main feature of the original image. More importantly, compared to the traditional CFAR, the proposed CP-CFAR is more suitable for GPU acceleration, which can improve the detection efficiency significantly. Experiments on the moving and stationary target acquisition and recognition SAR images with a size of 1478 × 1784 show that, compared with the traditional cell-averaging CFAR, two-parameter CFAR and their CPU, multithread CPU, and GPU acceleration modes, the proposed CP-CFAR with GPU acceleration can obtain the best detection performance with the highest acceleration ratio, and the operation time is less than 192 ms.

show abstract

Section: B Two-parameter Cfar Methodsmentioning

confidence: 99%

“…Compared with CFAR methods implemented with the sliding window technique, the computational burden of the proposed method is significantly reduced without loss of detection performance. An improved two-parameter CFAR method is proposed in [8]. The new two-parameter CFAR detector uses log-normal as the statistical model.…”

mentioning

confidence: 99%

SAR Target CFAR Detection Via GPU Parallel Operation

Cui

Quan

Cao

et al. 2018

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

View full text Add to dashboard Cite

show abstract

“…The development of non-coherent sliding window detectors, for operation in Burr distributed clutter, is now examined. A useful guide on such detectors is [1], while some more recent examples of development of such detectors include [28][29][30][31][32][33][34][35][36][37]. These detectors assume the existence of a series of independent and identically distributed clutter measurements, denoted Z 1 , Z 2 , .…”

Section: Development Of Non-coherent Detectorsmentioning

confidence: 99%

Burr Distribution for X-Band Maritime Surveillance Radar Clutter

Weinberg¹,

Tran²

2018

PIER B

View full text Add to dashboard Cite

Recent research has shown that the Pareto family of distributions provides suitable intensity models for high resolution X-band maritime surveillance radar clutter. In particular, the two parameter Pareto Type II model has been shown to fit the Australian Defence Science and Technology Group's medium to high grazing angle clutter returns very well. The Pareto Type II model is a special case of a Burr distributional model, which is a three parameter power law statistical model. Hence this paper begins by investigating the fitting of the Burr model to real data. Based upon these results a detailed study of the development of non-coherent sliding window detectors is justified, for operation in such clutter. Several different approaches will be applied to construct the decision rules. These include a transformation approach and direct adaptation of such detectors, designed for operation in exponentially distributed clutter, to the Burr clutter setting. In addition to this, the fact that the Burr distribution is invariant with respect to two of its distributional parameters allows specification of detection processes which have the constant false alarm rate property with respect to these model parameters. Performance analysis, in simulated clutter, of the derived detectors is then examined. This includes performance in the presence of interference and false alarm regulation during clutter power transitions. This is complemented by an application of the decision rules to target detection in real high resolution X-band maritime surveillance radar clutter.

show abstract

“…It has low computational complexity and adaptive threshold preserving constant false alarm probability by using the given background clutter distribution model and false alarm rate. However, complicated sea conditions cause the high nonhomogeneity of sea clutter in the SAR images, which affect the CFARbased automatic target detection [6]. Moreover, they often rely on a size parameter which may change for different images, providing, in some cases, too cluttered results.…”

Section: Introductionmentioning

confidence: 99%

Ship Detection in Sar Image Based on Information Geometry Method

Yang¹,

Yang²,

Zhang³

et al. 2019

PIER M

View full text Add to dashboard Cite

Aiming at the problem of high false alarm rate with respect to adaptive threshold in the ship detection from synthetic aperture radar (SAR) images, a novel strategy increasing robustness when using local adaptive threshold is proposed. In this article, we establish a fusion detection model based on a combination of the information geometry and surface geometry. Information geometry from a metric viewpoint can increase the contrast between targets and clutter in SAR image. Local surface feature gives a brief application of adaptive threshold method in ship detection from SAR images by means of the constant false-alarm-rate. Experiments indicate that the proposed geometry-based approach can effectively detect ship targets from complex background SAR images by using the method of fusion processing.

show abstract

An Adaptively Truncated Clutter-Statistics-Based Two-Parameter CFAR Detector in SAR Imagery

Cited by 56 publications

References 33 publications

SAR Target CFAR Detection Via GPU Parallel Operation

SAR Target CFAR Detection Via GPU Parallel Operation

Burr Distribution for X-Band Maritime Surveillance Radar Clutter

Ship Detection in Sar Image Based on Information Geometry Method

Contact Info

Product

Resources

About