This study assessed and mapped the aboveground tree carbon stock using very high-resolution satellite imagery (VHRS)—WorldView-2 in Barkot forest of Uttarakhand, India. The image was pan-sharpened to get the spectrally and spatially good-quality image. High-pass filter technique of pan-sharpening was found to be the best in this study. Object-based image analysis (OBIA) was carried out for image segmentation and classification. Multi-resolution image segmentation yielded 74% accuracy. The segmented image was classified into sal (Shorea robusta), teak (Tectona grandis) and shadow. The classification accuracy was found to be 83%. The relationship between crown projection area (CPA) and carbon was established in the field for both sal and teak trees. Using the relationship between CPA and carbon, the classified CPA map was converted to carbon stock of individual trees. Mean value of carbon stock per tree for sal was found to be 621 kg, whereas for teak it was 703 kg per tree. The study highlighted the utility of OBIA and VHRS imagery for mapping high-resolution carbon stock of forest.

中文翻译：

---

使用基于对象的图像分析 (OBIA) 技术对森林碳储量进行高分辨率制图

本研究使用位于印度北阿坎德邦 Barkot 森林的超高分辨率卫星图像 (VHRS) WorldView-2 评估并绘制了地上树木碳储量图。对图像进行全色锐化以获得光谱和空间质量良好的图像。全色锐化的高通滤波器技术被认为是本研究中最好的。基于对象的图像分析（OBIA）用于图像分割和分类。多分辨率图像分割产生了 74% 的准确率。分割后的图像被分为萨尔（Shorea Robusta）、柚木（Tectona grandis）和阴影。发现分类准确率为83%。在野外建立了盐树和柚木树的树冠投影面积 (CPA) 和碳之间的关系。利用 CPA 与碳的关系，分类的 CPA 地图被转换为单个树木的碳储量。Sal 每棵树的碳储量平均值为 621 公斤，而柚木每棵树的碳储量为 703 公斤。该研究强调了 OBIA 和 VHRS 图像在绘制高分辨率森林碳储量方面的效用。