论文部分内容阅读
文中构建一种对称Gabor小波滤波器(SGWF),能够对逆合成孔径雷达(ISAR)像中舰船进行识别。由Gabor小波滤波器构建的SGWF滤波器具有上下对称结构,对镜像翻转的图像滤波后仍具有镜像对称性。使用SGWF与奇异值分解(SVD)相结合来提取特征,使得该方法可以避免因多普勒变化导致的ISAR正像和倒像的检测问题。SGWF可在不同尺度和方向对图像进行滤波,充分反映图像的纹理特征和强度特征。在实验中,使用九艘民船的实测ISAR像进行识别,通过在成像过程加入高斯白噪声来检测算法对噪声的鲁棒性,并检验图像分块数量对算法性能的影响,以及检验训练样本数量对算法性能的影响,识别结果证明了算法的有效性。
In this paper, a symmetric Gabor wavelet filter (SGWF) is constructed to identify the ships in inverse synthetic aperture radar (ISAR) images. The SGWF filter constructed by the Gabor wavelet filter has a symmetrical structure up and down, and still has the mirror symmetry after the image with the reversed image is filtered. The combination of SGWF and singular value decomposition (SVD) to extract the features makes it possible to avoid the detection of ISAR eclipses and inversions due to Doppler shifts. SGWF filters images at different scales and orientations, fully reflecting the texture and intensity features of the image. In the experiment, the real ISAR images of nine civilian vessels were used to identify the robustness of the algorithm by adding Gaussian white noise during the imaging process, and the influence of the number of image blocks on the performance of the algorithm and the number of training samples The performance of the algorithm, the recognition results prove the effectiveness of the algorithm.