نوع مقاله : مقاله پژوهشی
نویسندگان
1 کارشناس ارشد سنجش از دور اداره مطالعات و مهندسی اداره کل منابع طبیعی و آبخیزداری استان فارس
2 کارشناس ارشد جنگلداری، اداره کل منابع طبیعی و آبخیزداری استان تهران
3 عضو هیئت علمی
چکیده
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
The capabilities of the UAVs and sensors that have been installed on them have led to an ever-increasing expansion of the equipment in various fields, including forest management. In this research, the accuracy of drone images in the investigation and identification of areas covered by Bux Tree dams in the forest of Sisangan was studied using classification algorithms. For this purpose, three pieces of circular sample with a radius of 60 meters and an area of 1.13 hectares were designed in the area. The density of Bux Tree dams and the bases of living and green were determined in them. Specimens of the sample were: Sample plot with domination of dieback Buxus bases, sample plot with overlapping of living and green bases, and sample plot with interstitial state in which the number of green dams were almost identical. Then, in each sample plot, at places where the bases of Buxus were located, there were 50 didactic points, as well as in the places where the living bases, the grass cover and the raspberries were located 50 points were recorded. After finishing the sampling and image processing and classification operations of all three plots of samples, using artificial neural network algorithms, the maximum likelihood and minimum distance were performed, and the results were compared with the overall accuracy evaluation indicators and the kappa coefficient was measured. Finally, the artificial neural network algorithm has the best results compared to the other two algorithms. This algorithm with a total accuracy of 97.47% and a kappa coefficient of 0.94 had the best results in the separation and detection of Buxus dieback in the sample plots with the domination of Buxus dieback. These results indicate that the use of emerging technologies such as UAVs in forest management and monitoring can have very promising results.
کلیدواژه‌ها [English]
)