Proper forest management needs quantitative and precise estimates of forest stands characteristics. Remotely sensed imageries, due to accurate and broad spatial information, has become a cost-effective tool in forest management. Classification of forest attributes and generation of thematic maps are among the common applications of remote sensing. The objective of this study was to optimize the density classification (number of trees per hectare) in forest stands using non-parametric KNN method in Pilambara, Shafarood watershed, west of Guilan province. This study showed that KNN method with k=6, as the optimum number of nearest neighbors, and Euclidian distance presented acceptable results with RMSE=228.58 (number of trees per hectare), relative RMSE=78.9% and correlation r=0.50 in mapping the stand densities in the study area. The accuracy rate and kappa coefficient of classified thematic map were 85.19% and 0.56, respectively. It is concluded that the KNN algorithm as a non-parametric method could classify the forest density properly.
Abedi, R., & Bonyad, A. (2017). Forest density classification using IRS satellite image and non-parametric KNN method. Forest and Wood Products, 69(4), 667-677. doi: 10.22059/jfwp.2017.60604
MLA
Roya Abedi; Amireslam Bonyad. "Forest density classification using IRS satellite image and non-parametric KNN method", Forest and Wood Products, 69, 4, 2017, 667-677. doi: 10.22059/jfwp.2017.60604
HARVARD
Abedi, R., Bonyad, A. (2017). 'Forest density classification using IRS satellite image and non-parametric KNN method', Forest and Wood Products, 69(4), pp. 667-677. doi: 10.22059/jfwp.2017.60604
VANCOUVER
Abedi, R., Bonyad, A. Forest density classification using IRS satellite image and non-parametric KNN method. Forest and Wood Products, 2017; 69(4): 667-677. doi: 10.22059/jfwp.2017.60604
)