Target Recognition for Multi-Aspect Sar Images With Fusion Strategies

Abstract

Abstract-Two fusion strategies for target recognition using multiaspect synthetic aperture radar (SAR) images are presented for recognizing ground vehicles in MSTAR database. Due to radar crosssection variability, the ability to discriminate between targets varies greatly with target aspect. Multi-aspect images of a given target are used to support recognition. In this paper, two fusion strategies for target recognition using multi-aspect SAR images are proposed, which are data fusion strategy and decision fusion strategy. The recognition performance sensitivity to the number of images and the aspect separations is analyzed for those two target recognition strategies. The two strategies are also compared with each other in probability of correct classification and operating efficiency. The experimental results indicate that if we have a small number of multi-aspect images of a target and the aspect separations between those images are proper, the probability of correct classification obtained by the two proposed strategies can be advanced significantly compared with that obtained by the method using single image.

HRR Profiles Time-Frequency Non-Negative Sparse Coding for Sar Target Classification

“…, the adjustable T -F center (t i , f i ) and associated variance α i can be obtained using FFT and the specific search procedure in [6]. The t i , f i , α i , and C i finally obtained give the solution of (2) and these four parameters completely describe one Gaussian T -F basis function at the i-th iteration.…”

“…The most straightforward approach is to use a SAR target image's intensity values to directly generate image features [4,5]. Another popular approach feature extraction technique is based on wavelet transform or multi-scale approach [6]. Alternatively, the features such as target region outline descriptors or ridges can also be extracted to classify targets [7,8].…”

Sar Target Classification Using Bayesian Compressive Sensing With Scattering Centers Features

“…In addition to target image, the shadows in SAR image is also used as features for SAR ATR [7]. Principle component analysis (PCA) and independent component analysis (ICA) are popular features extraction approaches for target classification [6,8,9]. SAR data, however, does not necessarily need to be processed as a single image.…”

“…The performance of a SAR ATR system is mainly decided by features extraction and classification algorithms [6]. A number of different feature extraction methods exist for the exploitation of SAR target images with respect to target classification.…”

Using Wave-Coefficients as Feature Vectors to Identify Aerospace Targets

“…The radar target identification can be realized using various target features, such as high resolution range profiles (HRRP) [1][2][3][4][5], synthetic aperture radar (SAR) [6][7][8][9] and inverse synthetic aperture radar (ISAR) [10,11] images, natural frequencies [12][13][14][15][16] and the RCS (radar cross section) data [17]. The HRRPs and RCSs of a target are easy to get, but are highly sensitive to the orientation of the targets.…”