Research on the Estimation of Chinese Fir Stand Volume Based on UAV-LiDAR Technology

Abstract

Chinese fir (Cunninghamia lanceolata) is the main fast-growing timber species in China, and studies of its stand volume are important for evaluating the effectiveness of forest management. However, it is difficult to accurately estimate stand volume from the perspective of a single tree due to the mutual concealment among Chinese fir trees. Therefore, in this study, we propose a method for identifying different forms of Chinese fir. The specific idea is to realize the accurate identification of a single Chinese fir tree, two Chinese fir trees, and three Chinese fir trees, and construct their respective stand volume estimation models to obtain an estimate of the forest stand volume. The key results are as follows: (1) the overall accuracy of recognition of different forms of Chinese fir is 79%, and the construction of different forms of Chinese fir units is beneficial for identifying forest trees; (2) a multiunit volume equation for different forms of Chinese fir is constructed; (3) based on predictions obtained with the constructed stand volume model, the difference between the estimated stand volume and the measured stand volume is small, and the average accuracy reaches 89.19%; and (4) compared to traditional volume estimation methods based on individual tree scale, the research method in this study shows a significant improvement (about 9.96%) in overall accuracy. In summary, this method can weaken the influence of erroneous individual tree segmentation on the accuracy of stand volume estimation, and can greatly reduce the working time of single tree segmentation to achieve the fast and accurate estimation of fir plantation stand volume.