为了对大视场范围内的目标进行 光谱成像，人们通常采用二维指向镜和面阵凝视光谱仪相结合 的技术途径。当用二维指向镜对近距离目标进行成像时，在不同方位 角与俯仰角下产生的像旋和畸变等都不同，从而给后续处理带来了 困难。因此需要先建立目标实际坐标系、探测器图像坐 标系和俯仰方位角三者之间的光学传递模型。根据该 模型即可得到透射畸变图像与真实图像之间的变换关系。然后通 过透射变换对透射畸变图像进行正视校正，便可对 大视场图像进行拼接。外场实验结果表明，该方法可行性 强，简单易行，稳定性好。

To achieve the spectral imaging of an object in a large field of view, the technology path which combines a two-dimensional pointing mirror with a staring spectrometer is usually used. However, when the two-dimensional pointing mirror is used to image close objects, the image rotation and distortion generated at different azimuths and pitch angles are different. Therefore, it is necessary to establish an optical transfer model among the actual object coordinate system, the detector image coordinate system and the azimuth and pitching angle firstly. According to the optical transfer model, the transformation relationship between the transmission distortion image and the real image can be obtained. Then, after the transmission distortion image is corrected by perspective transmission, image stitching can be implemented on the large field-of-view image. The experimental results show that the method is feasible, easily implemented and stable.