As one of the most challenging tasks in the field of SAR image processing, land target detection is also one of the focus of academic research. Synthetic aperture radar has the characteristics of all-sky, all-weather and cloud penetration, but it is difficult to detect the target in the image because of speckle noise and other problems. As a typical target, the detection of civil airport and civil aircraft has certain uniqueness. How to detect civil aircraft targets efficiently and accurately is one of the important issues and difficulties in the field of SAR image target interpretation. This paper focuses on the problem of airport and civil aircraft target detection in large scene high-resolution synthetic aperture radar, combined with deep learning method. Experimental results based on public data sets show that the proposed algorithm can accurately detect airport and aircraft targets in SAR images in various complex scenes, and the experimental results have strong robustness. As one of the most challenging tasks in the field of SAR image processing, land target detection is also one of the focus of academic research. Synthetic aperture radar has the characteristics of all-sky, all-weather and cloud penetration, but it is difficult to detect the target in the image because of speckle noise and other problems. As a typical target, the detection of civil airport and civil aircraft has certain uniqueness. How to detect civil aircraft targets efficiently and accurately is one of the important issues and difficulties in the field of SAR image target interpretation. This paper focuses on the problem of airport and civil aircraft target detection in large scene high-resolution synthetic aperture radar, combined with deep learning method. Experimental results based on public data sets show that the proposed algorithm can accurately detect airport and aircraft targets in SAR images in various complex scenes, and the experimental results have strong robustness.