Features for high resolution radar range profile based ship classification

Vessel classification using inverse synthetic aperture radar (ISAR) imagery is important because it can be used for maritime surveillance and has a high military value. We propose a vessel classification algorithm based on multifeature joint matching. We first utilize a preprocessing method to eliminate the vessel wakes and strong sea clutter, which interfere with feature extraction. In view of the different categories of vessels, we then propose a new twodimensional strong scattering points encoding (SSPE 2-D ) for vessel recognition. Furthermore, we modify the method to calculate the number of peaks in the range profile in order to obtain a more accurate result. The high-resolution ISAR images obtained as a result are used to verify the effectiveness of our method. We also compare our proposed method with three other classification methods, and show that the classification rate obtained using our technique is more accurate than that from each of the other methods. Our experiments also show that the preprocessing and the new encoding feature improve classification accuracy.

show abstract

“…16, the length of the feature was found to be the most important factor in classification accuracy. The length of the vessel.…”

Section: Feature Extractionmentioning

confidence: 98%

Hierarchical vessel classifier based on multifeature joint matching for high-resolution inverse synthetic aperture radar images

Zhao

Wang

et al. 2014

J. Appl. Remote Sens

View full text Add to dashboard Cite

show abstract

“…Some basic algorithms have used ship length to coarsely classify ships, for example, distinguish a cargo from a fishing vessel [19]. The correct retrieval of this parameter is not easy as accuracy depends mainly on the ship's size compared with the available resolution, its orientation, and its motion history [20], [21].…”

Section: A Vessel Motion Effects On Ship Length Retrieving 1) Theorymentioning

confidence: 99%

On the Usage of GRECOSAR, an Orbital Polarimetric SAR Simulator of Complex Targets, to Vessel Classification Studies

Margarit

Mallorquí

Rius

et al. 2006

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Abstract-This paper presents a synthetic aperture radar (SAR) simulator that is able to generate polarimetric SAR (POLSAR) and polarimetric inverse SAR data of complex targets. It solves the electromagnetic problem via high-frequency approximations, such as physical optics and the physical theory of diffraction, with notable computational efficiency. In principle, any orbital monostatic sensor working at any band, resolution, and operating mode can be modeled. To make simulations more realistic, the target's bearing and speed are considered, and for the particular case of vessels, even the translational and rotational movements induced by the sea state. All these capabilities make the simulator a powerful tool for supplying large amounts of data with precise scenario information and for testing future sensor configurations. In this paper, the usefulness of the simulator on vessel classification studies is assessed. Several simulated polarimetric images are presented to analyze the potentialities of coherent target decompositions for classifying complex geometries, thus basing an operational algorithm. The limitations highlighted by the results suggest that other approaches, like POLSAR interferometry, should be explored.

show abstract

“…The quality of the extracted features determines the performance of target recognition. Therefore, many scholars [ 3 , 6 , 10 , 11 , 12 , 13 , 14 , 15 ] have spent a lot of effort studying the methods of HRRP feature extraction. In [ 3 ], the principal component analysis (PCA) subspace model is utilized to minimize the reconstruction error.…”

Section: Introductionmentioning

confidence: 99%

“…The multitask learning truncated stick-breaking hidden Markov model (MTL TSB-HMM) proposed in [ 6 ] is used to characterize the fast fourier transform (FFT) magnitude features of HRRP. Some other researchers [ 10 , 11 ] have used complicated statistic models to extract features from HRRP that have specific physical meaning, such as the target size, the center of gravity, the number of peaks, and so on. By using the RELAX and other super-resolution algorithms, the precise location and intensity information of radar HRRP scatterers can be extracted [ 12 , 13 ].…”

Section: Introductionmentioning

confidence: 99%

Radar HRRP Target Recognition Based on Stacked Autoencoder and Extreme Learning Machine

Zhao

Liu

Huang

et al. 2018

Sensors

View full text Add to dashboard Cite

A novel radar high-resolution range profile (HRRP) target recognition method based on a stacked autoencoder (SAE) and extreme learning machine (ELM) is presented in this paper. As a key component of deep structure, the SAE does not only learn features by making use of data, it also obtains feature expressions at different levels of data. However, with the deep structure, it is hard to achieve good generalization performance with a fast learning speed. ELM, as a new learning algorithm for single hidden layer feedforward neural networks (SLFNs), has attracted great interest from various fields for its fast learning speed and good generalization performance. However, ELM needs more hidden nodes than conventional tuning-based learning algorithms due to the random set of input weights and hidden biases. In addition, the existing ELM methods cannot utilize the class information of targets well. To solve this problem, a regularized ELM method based on the class information of the target is proposed. In this paper, SAE and the regularized ELM are combined to make full use of their advantages and make up for each of their shortcomings. The effectiveness of the proposed method is demonstrated by experiments with measured radar HRRP data. The experimental results show that the proposed method can achieve good performance in the two aspects of real-time and accuracy, especially when only a few training samples are available.

show abstract

Features for high resolution radar range profile based ship classification

Cited by 23 publications

References 6 publications

Hierarchical vessel classifier based on multifeature joint matching for high-resolution inverse synthetic aperture radar images

Hierarchical vessel classifier based on multifeature joint matching for high-resolution inverse synthetic aperture radar images

On the Usage of GRECOSAR, an Orbital Polarimetric SAR Simulator of Complex Targets, to Vessel Classification Studies

Radar HRRP Target Recognition Based on Stacked Autoencoder and Extreme Learning Machine

Contact Info

Product

Resources

About