Accuracy assessment of tree crown detection using local maxima and multi-resolution segmentation

8th International Symposium of the Digital Earth

地球科学;计算机科学

Khalid, N.^1 ; Hamid, J.R.A.^1 ; Latif, Z.A.^1

Centre for Geospatial Technology (CGT), Faculty of Architecture, Planning and Surveying, Universiti Teknologi MARA, Selangor, 40450 Shah Alam, Malaysia^1

关键词: Accuracy assessment;    Forest ecosystem;    Forest reserves;    Multi resolutions;    Multi-resolution image segmentations;    Parameter measurement;    Remote sensing technology;    Tree distribution;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/18/1/012023/pdf
DOI  :  10.1088/1755-1315/18/1/012023

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Diversity of trees forms an important component in the forest ecosystems and needs proper inventories to assist the forest personnel in their daily activities. However, tree parameter measurements are often constrained by physical inaccessibility to site locations, high costs, and time. With the advancement in remote sensing technology, such as the provision of higher spatial and spectral resolution of imagery, a number of developed algorithms fulfil the needs of accurate tree inventories information in a cost effective and timely manner over larger forest areas. This study intends to generate tree distribution map in Ampang Forest Reserve using the Local Maxima and Multi-Resolution image segmentation algorithm. The utilization of recent worldview-2 imagery with Local Maxima and Multi-Resolution image segmentation proves to be capable of detecting and delineating the tree crown in its accurate standing position.