Individual Populus euphratica tree detection in sparse desert forests based on constrained 2D bin packing

Detecting individual Populus euphratica (P. euphratica) trees in desert forest areas is crucial for monitoring their ecophysiological characteristics and ecological conservation. However, the presence of the spectral-similar Tamarix chinensis (T. chinensis) in the habitats, along with the densely overlapping crowns in clustered P. euphratica, presents a challenge for the task. This article proposes a method to detect individual P. euphratica in very high spatial resolution (VHR) images. First, the deep learning-based semantic segmentation model is used to differentiate between P. euphratica and T. chinensis. Second, the individual tree detection is converted into a constrained 2-D bin packing model and solved by a heuristic template matching and filling algorithm. The experimental data consist of a WorldView-2 image capturing sparse desert forests of the lower reaches of the Tarim River. The 22 296 individual P. euphratica trees were detected, achieving F1-scores of 0.885, 0.869, and 0.902 on three datasets with varying difficulty levels. Furthermore, experiments were conducted to compare with other methods, and the results showed that the proposed method achieved the best performance on all three datasets. The proposed method can be applied to map the distribution of individual P. euphratica trees in sparse desert forests and can provide methodological references for similar tasks related to individual tree detection in natural forests.