Appl Opt. 2020 Apr 1;59(10):C70-C77. doi: 10.1364/AO.382081.
The spatial resolution of an observation from a geostationary orbiting satellite is usually too coarse to track small scale macroalgae blooms. For macroalgae mapping to benefit from a geostationary orbit's staring monitoring and frequent revisit intervals, we introduced a super-resolution method that reconstructs a high-resolution (HR) image of a region from a sequence of raw geostationary low-resolution images of the same region. We tested our method with GF-4 images at 50 m spatial resolution and demonstrated that the spatial resolution increased to 25 m. In addition, the derived HR image had better image quality characterized by a higher signal-to-noise ratio, clarity, and contrast. The increased spatial resolution and improved image quality improved our ability to distinguish macroalgae patches from the surrounding waters, especially tiny patches of macroalgae, and to precisely delineate the patch boundaries. Lastly, we more accurately estimated the areal coverage of the patches by reducing underestimation of the coverage of tiny patches and overestimation of the coverage of large patches.
从地球静止轨道卫星进行观测的空间分辨率通常太粗糙，无法跟踪小尺度大型藻类浮。为了使大型藻类测绘受益于地球静止轨道的凝视监测和频繁的重访间隔，我们引入了一种超分辨率方法，该方法可以从同一区域的一系列原始地球静止低分辨率图像中重建该区域的高分辨率 (HR) 图像。我们使用空间分辨率为 50 米的 GF-4 图像对我们的方法进行了测试，并证明空间分辨率提高到了 25 米。此外，所得 HR 图像具有更好的图像质量，其特点是更高的信噪比、清晰度和对比度。更高的空间分辨率和改进的图像质量提高了我们区分大型藻类斑块与周围水域的能力，特别是区分细小的大型藻类斑块的能力，并能够更精确地划定斑块边界。最后，我们通过减少对小斑块覆盖范围的低估和对大斑块覆盖范围的高估，更准确地估计了斑块的面积。
