object-based

Forest canopy gaps play an important role in forest dynamics. Unmanned aerial vehicle (UAV) data provide demonstrated capacity to systematically and accurately detect and map canopy gaps and have been considered as an alternative way to describe the forest stands. This study aims to extract canopy gaps using UAV data and compare the performance of different canopy gap extraction methods in a part of the replanted forest in the Arab Dagh Region, Golestan Province, Iran.

After the acquisition of UAV images and initial preprocessing, the digital terrain model (DTM), digital surface model (DSM), Canopy height model (CHM), and orthophoto mosaic were produced. CHM classification performs to extract forest gaps by different methods of height thresholding on CHM, CHM slope thresholding, and object-based classification. For performance evaluation of used methods and accuracy assessment of the canopy gap maps, the central position and boundary of some gaps were measured by DGPS. Finally, the point and polygon base accuracy of delineated gaps were assessed for each of the methods..

The results of the point accuracy assessment showed that the canopy gap map obtained by object-based classification method with applying the support vector machine (SVM) algorithm with 99% overall accuracy and 0.98 kappa coefficient had the best performance compared to other algorithms and methods. About area accuracy assessment, the best match between delineated gaps and ground truth polygons was achieved by using 3 m height thresholding.

The results showed that with aerial images of the UAV and its outputs, as well as the use of automated methods, the map of the canopy gap can be accurately extracted. Of course, the degree of accuracy depends on several factors such as the type of drone and cameras used, flight parameters and so on. Given the results, it is hoped that this approach will gradually be used as a cheap and accurate method in forest surveying.

Canopy gap delineation is essential for achieving a better comprehension of forest structure. This study aims to (a) extract canopy gaps using UAV data and (b) compare the performance of different canopy gap extraction methods in a managed stand in the northeast of Iran. A canopy height model (CHM) was produced by subtracting LIDAR digital terrain model from the UAV digital surface model. CHM classification performs to extract gaps by thresholding CHM (fixed height and CHM slope and relative height thresholds) and object-based classification on the UAV CHM and orthophoto. Ground truth is produced in the point and polygon forms through field measurements and visual interpretation of the UAV orthophoto. The geometry of the canopy gaps (Area, perimeter, and shape complexity) was calculated. Finally, the point and polygon base accuracy of delineated gaps assess for each of the methods. Point accuracy assessment suggests that 60% CHM slope produces the highest overall accuracy and Kappa coefficient of 91.7% and 0.874, respectively. About area accuracy assessment, the best match between delineated gaps and ground truth polygons was achieved by using relative height and 60% CHM slope thresholds. The lowest mean errors of GSCI produced by 70% CHM slope (0.15). Moreover, object-based classification showed the lowest mean error of area (33.76 m2) and perimeter (16.80 m). In conclusion, while area accuracy is considered the best fit of the delineated gap's geometry is gained by the object-based classification.

Unmanned aerial vehicles (UAVs) images have high spatial resolution. They are a valuable source of information for mapping land cover and thematic information, particularly in the identification of tree species. The aim of this study was to investigate the capability of drone images and the base object method for detecting tree species in the Hyrcanian forests. For this purpose, part of an area in parcel 24 of district one in Mazandaran Darabkola forest was selected. The ground truth map was prepared through accurate recording with geographic coordinate’s algorithm using distance and azimuth in MATLAB software. Proper processing was done on the images and classification performed on images at three flight height; 55, 75 and 100 meters in two categories of one-step and hierarchical classifications. In object-based classification, the nearest neighbor method was used to classify three species. The accuracy of the maps derived from classifications was evaluated using 50% of the ground truth map. The results showed that the map of the hierarchical classification by the object based method at a flight height of 55 meters has the best ability to detect tree species in the three heights. These comparisons showed Kappa's coefficient of 0.81 accuracy of tree species classification in 55-meter height by UAV.