空间目标的ISAR成像及轮廓特征提取

空间目标的逆合成孔径雷达（ISAR）成像由于受自身遮挡及噪声干扰等影响，导致生成的ISAR像难以直接进行图像分析及目标识别。由此，以空间目标的ISAR成像建模为基础对ISAR像处理及特征提取展开了研究。首先，分别建立了目标卫星的ISAR成像模型、ISAR信号模型及ISAR图像函数提取模型，并经过旁瓣抑制与相干斑滤波等初步处理得到了目标卫星的ISAR像。其次，采用Otsu算法、canny算子及Hough变换使卫星旋转至最长轴平行于像平面横轴，通过闭运算填充卫星内部孔洞，去除外部孤立噪声，并基于连通域思想分割出卫星所在子区域，实现了卫星的轮廓提取。所设计的图像处理算法能有效改善ISAR像质量，提取的卫星轮廓线能较好地勾勒出目标卫星的外形结构，为进一步展开卫星的识别工作奠定了重要基础。 

ISAR imaging and contour feature extraction of space targets

The inverse synthetic aperture radar (ISAR) imaging of the space target is affected by target's own occlusion and noise interference, which makes the generated ISAR image difficult to be directly used for image analysis and target recognition. Therefore, the ISAR image processing and feature extraction are studied based on the ISAR imaging model of the space target. Firstly, the ISAR imaging model, ISAR signal model and ISAR image extraction model of the target satellite are established respectively, and the ISAR image of the target satellite is obtained through preliminary processing such as sidelobe suppression and speckle filtering. Secondly, the Otsu algorithm, the canny operator and the Hough transform are used to rotate the satellite to the longest axis parallel to the horizontal axis of the image plane. The closed operation method is used to fill the internal cavity of the satellite and remove the external isolated noise region. Based on the connected domain idea, the sub-area where the satellite is located is segmented, and the contour extraction of the satellite is realized. The designed image processing algorithm can effectively improve the ISAR image quality, and the extracted satellite contour can well outline the shape of the target satellite, which lays an important foundation for further satellite identification.