PROJECT TITLE :
Focal-Test-Based Spatial Decision Tree Learning
Given learning samples from a raster data set, spatial call tree learning aims to find a decision tree classifier that minimizes classification errors with salt-and-pepper noise. The problem has vital societal applications like land cover classification for natural resource management. But, the problem is difficult due to the fact that learning samples show spatial autocorrelation in school labels, rather than being independently identically distributed. Related work depends on native tests (i.e., testing feature data of a location) and cannot adequately model the spatial autocorrelation result, resulting in salt-and-pepper noise. In contrast, we recently proposed a focal-check-based spatial decision tree (FTSDT), in which the tree traversal direction of a sample is predicated on each native and focal (neighborhood) information. Preliminary results showed that FTSDT reduces classification errors and salt-and-pepper noise. This paper extends our recent work by introducing a brand new focal take a look at approach with adaptive neighborhoods that avoids over-smoothing in wedge-shaped areas. We tend to additionally conduct computational refinement on the FTSDT training algorithm by reusing focal values across candidate thresholds. Theoretical analysis shows that the refined coaching algorithm is correct and additional scalable. Experiment results on real world information sets show that new FTSDT with adaptive neighborhoods improves classification accuracy, and that our computational refinement significantly reduces training time.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here