海天线检测是红外航拍图像的一项重要研究内容。深入探讨海天线检测问题，给出了一种实用的检测算法。首先，对获取的红外图像进行梯度运算，采用Ostu 阈值分割方法，获取海天边缘信息；然后根据Hough 变换原理，完成海天线的检测。此外，通过随机点采样提高了海天线检测的速率，通过多线段拟合提高了检测的准确率。实验结果表明，本算法可以准确、高效地检测出海天线，为下一步研究奠定了基础。

Sea-sky-line detection is an important research content of infrared aerial image. The paper focuses on the problem of sea-sky-line detection and proposes a practical detection method. Firstly, the paper uses the way of calculating the gradient of the obtained infrared image and segmenting the image based on OSTU to get the edge information. Then the paper detects the sea-sky-line based on the Hough Transform. In addition, random sampling is used to improve detection speed and multi-line fitting is adopted to improve detection accuracy. Experimental results show that the algorithm can detect the sea-sky-line accurately and efficiently, which lays a foundation for further study.