The rising global demand for palm oil necessitates more efficient plantation management, yet the accuracy and effectiveness of monitoring remain limited. Photogrammetric technology provides a precise solution for tree counting and height estimation, traditionally relying on aerial imagery and ground control points (GCPs). However, advancements in direct georeferencing photogrammetry enable accurate tree detection without GCPs. This study validates automatic segmentation using direct georeferencing aerial photographs, comparing results with manual methods. A template matching algorithm was applied for segmentation, while tree height estimation was derived from a canopy height model (CHM) using digital elevation model (DEM) data. The DEM was processed into a digital terrain model (DTM) and digital surface model (DSM) to generate CHM values. The eCognition oil palm application (OPA) software detected trees across homogeneous, semi-homogeneous, and heterogeneous areas. Accuracy results showed high precision (96.2%–98.8%), recall (99.4%– 100.0%), and F1-scores (98.0%–99.0%) across all areas. CHM-derived height estimates averaged 5.1 ± 1.8, 5.2 ± 2.0, and 6.9 ± 3.5 m, respectively. The results for each sample were consistent with the characteristics of the area, as seen from the differences in standard deviation, which is an indicator of the degree of variation in tree height. These findings highlight direct georeferencing photogrammetry as an effective, scalable approach for accurate tree counting and height estimation, supporting sustainable oil palm management.