2015-2017年生长季大兴安岭区域森林密度图处理过程示例数据

该研究主要是通过获取大量的高分辨率GF-2星载遥感数据，进行大兴安岭大尺度区域森林株数密度制图，并分析国产高分辨率光学遥感卫星数据GF-2在大尺度森林株数密度制图过程中所面临的几个关键问题，提出改正方法，进而提高区域大尺度森林株数密度制图的精度。研究拟提出一种局部自适应阈值法，该算法考虑大数据量及不同的场景、时相等条件时的算法稳定性和可移植性，预期可较好的去除背景信息，并结合局部最大值法获取森林株数密度。另外，通过分析影响森林株数密度估算的几个关键问题，即云覆盖的识别及补偿、地形的改正、太阳高度角变化对密度识别的影响以及多期数据之间的定标，并给出改正参数，对模型进行优化，最终得到区域尺度上的森林株数密度信息。该研究可以克服低分辨率数据大尺度森林株数密度估算不能反映森林的空间结构信息这一短板，而且从机理上说明森林株数密度的算法依据，能有效节约林业调查中的人力物力和时间成本，并对影响森林株数密度估算精度所面临的几个关键问题给予解释，对于大兴安岭区域尺度上森林株数密度精确制图具有现实意义，并为未来的大批量高分卫星数据在森林株数密度研究方面提供新思路和新借鉴。

This study primarily acquires massive high-resolution Gaofen-2 (GF-2) satellite-borne remote sensing data to conduct large-scale forest stem density mapping in the Daxing'an Mountains. It analyzes several key issues faced by domestically produced high-resolution optical remote sensing satellite GF-2 during large-scale forest stem density mapping, proposes corresponding correction methods, and thereby improves the accuracy of regional large-scale forest stem density mapping.
This study intends to propose a local adaptive threshold method, which considers algorithm stability and portability under conditions of large data volume, diverse scenes and phenological phases. The method is expected to effectively remove background information, and combine with the local maximum method to obtain forest stem density values.
In addition, by analyzing several key issues affecting forest stem density estimation, namely cloud cover identification and compensation, topographic correction, the impact of solar elevation angle variation on density identification, and calibration among multi-temporal datasets, and providing correction parameters to optimize the model, this study ultimately obtains forest stem density information at the regional scale.
This study overcomes the limitation that large-scale forest stem density estimation using low-resolution data fails to reflect the spatial structure information of forests. It also mechanistically explains the theoretical basis of the forest stem density algorithm, effectively saves human, material and time costs in forestry inventory, and interprets several key issues affecting the accuracy of forest stem density estimation. This research has practical significance for accurate forest stem density mapping at the regional scale of the Daxing'an Mountains, and provides new ideas and references for future research on forest stem density using massive high-resolution satellite data.

搜集汇总

背景与挑战

背景概述

该数据集是2015-2017年生长季大兴安岭区域森林密度图处理过程的示例数据，基于国产GF-2高分辨率卫星遥感数据（分辨率1米），旨在通过局部自适应阈值方法优化大规模森林数量密度制图，解决云覆盖、地形校正等关键问题以提高精度。数据集提供了栅格格式的示例，用于支持森林资源调查和区域尺度密度制图研究。