Fast two‐dimensional subset censored CFAR method for multiple objects detection from acoustic image

Gao, Jue; Li, Haisen; Chen, Baowei; Zhou, Tian; Xu, Chao; Du, Weidong

doi:10.1049/iet-rsn.2016.0322

Cited by 13 publications

(8 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, background noise distribution estimation is very important for target detection. Compared with the lognormal distribution and K-distribution, the Weibull distribution provides a trade-off between the accuracy of modeling and computational cost [2]. For these reasons, the Weibull distribution is more suitable to describe the statistical properties of this type of background noise.…”

Section: Background Noise Distribution Estimation Of Multibeammentioning

confidence: 99%

“…In recent years, with the constant exploration of oceans, multibeam sonar (MBS) is being applied in many scenarios, such as underwater environment mapping [1], underwater target detection [2], and underwater terrain-aided navigation [3]. However, noise interference is strong in water, which makes it more difficult to extract useful information from the images.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A SIFT‐Like Feature Detector and Descriptor for Multibeam Sonar Imaging

Zhang

Zhou

et al. 2021

Journal of Sensors

Self Cite

View full text Add to dashboard Cite

Multibeam imaging sonar has become an increasingly important tool in the field of underwater object detection and description. In recent years, the scale-invariant feature transform (SIFT) algorithm has been widely adopted to obtain stable features of objects in sonar images but does not perform well on multibeam sonar images due to its sensitivity to speckle noise. In this paper, we introduce MBS-SIFT, a SIFT-like feature detector and descriptor for multibeam sonar images. This algorithm contains a feature detector followed by a local feature descriptor. A new gradient definition robust to speckle noise is presented to detect extrema in scale space, and then, interest points are filtered and located. It is also used to assign orientation and generate descriptors of interest points. Simulations and experiments demonstrate that the proposed method can capture features of underwater objects more accurately than existing approaches.

show abstract

Section: Background Noise Distribution Estimation Of Multibeammentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A SIFT‐Like Feature Detector and Descriptor for Multibeam Sonar Imaging

Zhang

Zhou

et al. 2021

Journal of Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…The constant false alarm rate (CFAR) detector is regarded as the classical radar detector for automotive radar [4]. To deal with the impact caused by typical problems of CFAR such as the masking effects and computation complexity, some modified methods [5][6][7] based on cell averaging (CA)-CFAR or ordered statistic (OS)-CFAR are proposed. However, when applied to the complex scenes in autonomous driving, for example, in a high-density clutter environment such as the guardrails, tunnels, and soundproof walls, the performance of CFAR detector decreases dramatically as the number of clutter points increase a lot [8].…”

Section: Introductionmentioning

confidence: 99%

A New Automotive Radar 4D Point Clouds Detector by Using Deep Learning

Cheng

Chen

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

The millimeter-wave radar, as an important sensor, is widely used in autonomous driving. In recent years, to meet the requirement of high level autonomous driving applications, attentions have been paid to generate high-quality radar point clouds. However, in the complex roadway environment, the weaknesses of classical radar detectors are exposed, such as too much clutter points and sparse valid point clouds. Therefore, in this paper, we propose a new automotive radar detector based on deep learning using the spatial distribution feature of the real targets, in order to improve the performance of automotive radar detector in the real-world driving scene. Besides, aiming at the lack of radar data labels, we propose an autonomous labeling method by using synchronized Lidar data. Finally, we evaluate the detector on data collected in the real-world roadway scene and the result shows that the proposed radar detector out-performs the classical radar detectors in suppressing the clutter and generating denser point clouds.

show abstract

“…On the other hand, the constant false alarm rate (CFAR) represents an adaptive technique able to perform accurate and robust target detection. This technique is commonly used in radar technology for detecting moving objects [9, 10] as well also in sonar technology applied in acoustic images from different sonar devices for multi‐target detection [11–13], underwater pipeline detection on the seafloor [14, 15], acoustic segmentation of several types of regions [16], among others. CFAR calculates an adaptive detection threshold from interference power values to maintain an expected false alarm probability [9].…”

Section: Introductionmentioning

confidence: 99%

“…The main drawback of OS‐CFAR is its computational effort because sorting is a time‐consuming task. This computational effort prevents its use in real‐time applications [11], and therefore the utility of OS‐CFAR technique decreases. In addition, a two‐dimensional (2D) sliding window is necessary to consider more contextual information and hence to improve detections.…”

Section: Introductionmentioning

confidence: 99%

Efficient approach for OS‐CFAR 2D technique using distributive histograms and breakdown point optimal concept applied to acoustic images

Villar

Menna

Torcida³

et al. 2019

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

In this work, a new approach to improve the algorithmic efficiency of the Order Statistic-Constant False Alarm Rate (OS-CFAR) applied in two dimensions (2D) is presented. OS-CFAR is widely used in radar technology for detecting moving objects as well as in sonar technology for the relevant areas of segmentation and multi-target detection on the seafloor. OS-CFAR rank orders the samples obtained from a sliding window around a test cell to select a representative sample that is used to calculate an adaptive detection threshold maintaining a false alarm probability. Then, the test cell is evaluated to determine the presence or absence of a target based on the calculated threshold. The rank orders allows that OS-CFAR technique to be more robust in multi-target situations and less sensitive than other methods to the presence of the speckle noise, but requires higher computational effort. This is the bottleneck of the technique. Consequently, the contribution of this work is to improve the OS-CFAR 2D with the distributive histograms and the optimal breakdown point optimal concept, mainly from the standpoint of efficient computation. In this way, the OS-CFAR 2D on-line computation was improved, by means of speeding up the samples sorting problem through the improvement in the calculus of the statistics order. The theoretical algorithm analysis is presented to demonstrate the improvement of this approach. Also, this novel efficient OS-CFAR 2D was contrasted experimentally on acoustic images.

show abstract

Fast two‐dimensional subset censored CFAR method for multiple objects detection from acoustic image

Cited by 13 publications

References 28 publications

A SIFT‐Like Feature Detector and Descriptor for Multibeam Sonar Imaging

A SIFT‐Like Feature Detector and Descriptor for Multibeam Sonar Imaging

A New Automotive Radar 4D Point Clouds Detector by Using Deep Learning

Efficient approach for OS‐CFAR 2D technique using distributive histograms and breakdown point optimal concept applied to acoustic images

Contact Info

Product

Resources

About