Підбір незалежних змінних для класифікації лісового покриву за сезонними мозаїками Landsat

В. Миронюк

Анотація


Abstract. Development of the methods for mapping forest cover using
satellite images is an important support of the national forest inventory of Ukraine.
The objective of the research was to investigate the variable selection procedure
in a context of forest mask creation for the territory of lowland plains of Ukraine
using Landsat 8 OLI seasonal composited mosaics and Random Forest (RF)
algorithm. The study is based on reference dataset that includes more than 4700
sampling points. For visual interpretation of each sampling point, we used free
images of very high spatial resolution available from Google and Bing Maps
services. All images were filtered for the study area and combined in the form of four seasonal composited mosaics for the next periods: year, summer, autumn and April-October as it was described by (Roy et all., 2010). The accuracy of the 

classification models were tested by means of OBB (out-of-bag) error and variable importance were analyzed using %IncMSE provided by the randomForest algorithm for R. Using these measures, the conclusion was drawn that classification which incorporates all four mosaics had the higher accuracy followed by classification of April-October mosaics. The error obtained during classification of other mosaics were significantly higher. It also was found that inclusion of latitude and longitude in the list of predictors tends to increase the accuracy of classification of landcover types for the study territory. We used forward stepwise selection algorithm to analyze the accuracy of classification with different number of predictors (from 2 to 53). Finally, we concluded that the most accurate is the classification which incorporates 36 most important variables including longitude and latitude.
Keywords: Landsat 8 OLI, seasonal composited mosaics, forest mask,
Random Forest.


Повний текст:

PDF

Посилання


Baig M. H. A. Derivation of a tasselled cap transformation based on

Landsat 8 at-satellite reflectance / M. H. A. Baig, L. F. Zhang, T. Shuai,

Q. X. Tong // Remote Sensing Letters. – 2014. – Vol. 5 (5). – P. 423–431.

Breiman L. Random forests / L. Breiman // Machine Learning. – 2001. –Vol. 45 (1). – P. 5–32.

Chrysafis I. Estimating Mediterranean forest parameters using multi

seasonal Landsat 8 OLI imagery and an ensemble learning method /

I. Chrysafis, G. Mallinis, I. Gitas, M. Tsakiri-Strati // Remote Sensing of

Environment. – 2017. – Vol. 199. – P. 154–166.

Continuous fields of land cover for the conterminous United States using Landsat data: first results from the Web-Enabled Landsat Data (WELD) project / M. C. Hansen, A. Egorov, D. P. Roy et al. // Remote Sensing Letters. – 2011. – Vol. 2 (4). – P. 279–288.

Demonstration of Percent Tree Cover Mapping Using Landsat Analysis

RМeady Data (ARD) and Sensitivity with Respect to Landsat ARD

Processing Level / A. V. Egorov, D. P. Roy, H. K. Zhang, M. C. Hansen,

A. Kommareddy // Remote Sensing. – 2018. – Vol. 10. – P. 209.

Forest Monitoring Using Landsat Time Series Data: A Review / A. Banskota, N. Kayastha, M. J. Falkowski et al. // Canadian Journal of Remote Sensing. – 2014. – Vol. 40 (5). – P. 362–384.

Genuer R. Variable selection using random forests / R. Genuer, J. M. Poggi, C. Tuleau-Malot // Pattern Recognition Letters. – 2010. – Vol. 31 (14). – P. 2225–2236.

Hansen M. C. A review of large area monitoring of land cover change using Landsat data / M. C. Hansen, T. R. Loveland // Remote Sensing of

Environment. – 2012. – Vol. 122. – P. 66–74.

Monitoring conterminous United States (CONUS) land cover change with Web-Enabled Landsat Data (WELD) / M. C. Hansen, A. Egorov,


Метрики статей

Завантаження метрик ...

Metrics powered by PLOS ALM

Посилання

  • Поки немає зовнішніх посилань.