Estimating the parameters of forest inventory using machine learning and the reduction of remote sensing features

Locally computed statistics of image texture and a case-based reasoning (CBR) system were evaluated for mapping of forest attributes. Cluster analysis was preferred to regression models, as a pre-selection method of features. The best stand-based accuracy using satellite sensor images was 74.64 m−3 ha−1 (36%) RMSE for stand volume, 1.98 m−3 ha−1 a−1 (49%) for annual increase in stand volume, where κ = 0.23 for stand growth classes and κ = 0.41 for dominant tree species in stands. The top pixel-based accuracy using orthophotos was 76.54 m−3 ha−1 (41%) RMSE for stand volume, 1.87 m−3 ha−1 a−1 (44%) for annual increase in stand volume, where κ = 0.24 for stand growth classes and κ = 0.38 for dominant tree species in stands. Mean saturation in 30 m radius was the most useful feature when orthophotos were used, and standard deviation of Landsat ETM 6.2 values in 80 m radius was the best when satellite sensor images were used. The most valuable feature components (radii, channels and local statistics) for orthophotos were: 30 m kernel radius, lightness and the mean of pixel values; for satellite sensor images: 80 m kernel radius, near-infrared channel (ETM 4) and the mean of pixel values. Locally computed statistics.