在实现无人机航空相机的高精度标定中,模板图像中角点提取的精度是至关重要的因素。为了实现高精度角点的提取,以及亚像素角点像素的修正,提出了基于Hessian矩阵的X型角点形状算子的亚像素角点提取方法。实验分析验证,该方法提取精度较高,提取过程简单,计算量小,大大简化了标定过程。 In achieving high-precision calibration of a UAV aerial camera, the accuracy of corner extraction in a template image is a crucial factor. In order to extract the high-precision corners and to correct the pixels at the sub-pixel corners, a sub-pixel corner extraction method based on the Hessian matrix is ​​proposed. Experimental analysis shows that this method has high extraction accuracy, simple extraction process and small amount of calculation, which greatly simplifies the calibration process.