连续校正在采用IDGCAL校准方法获取的宽场成像中的应用
Application of continuous corrections in widefield imaging obtained with the IDGCAL calibration method
AI总结：
针对射电天文成像的IDG-CAL校准算法，提出两项改进以降低高阶A项计算成本并优化A项基函数，为深场成像的A项估计提供优化方案。
AI中文摘要：
图像域网格化（IDG）是评估A投影的高效方法，A投影是用于校正射电天文成像中方向相关效应（DDEs）的方法。校正项即A项需通过校准从观测数据中估计得到。IDG-CAL是一种校准算法，它以迭代方式采用IDG算法直接估计A项。低阶A项的初步结果颇具前景，但对于更深的图像则需要高阶A项。本文针对IDG-CAL提出两项改进：1）降低由更多参数描述的A项的计算成本；2）基于增益变化的随机模型和模型图像，优化描述A项的基函数集合。
英文摘要：
Image Domain Gridding (IDG) is an efficient method to evaluate A-projection, a method for correcting for direction dependent effects (DDEs) in radio astronomical imaging. The corrections i.e. the A-terms need to be estimated from the observed data through calibration. IDG-CAL is a calibration algorithm that employs the IDG algorithm in an iterative fashion to estimate A-terms directly. Initial results for low order A-terms are promising, however for deeper images higher order A-terms are needed. Here two improvements to IDG-CAL are proposed to 1) reduce the computational cost of A-terms described by more parameters and 2) optimize the set of basis functions describing the A-terms based on a stochastic model of the gain variations and the model image.
