3D Iterative Spatiotemporal Filtering for Classification of Multitemporal Satellite Data Sets

TL;DR

This paper introduces a 3D iterative spatiotemporal filtering method that enhances multitemporal satellite data classification accuracy by refining probability maps, addressing inconsistencies caused by varying acquisition conditions.

Contribution

The paper proposes a novel 3D spatiotemporal filtering approach that improves classification consistency and accuracy in multitemporal satellite data analysis.

Findings

Improved classification accuracy by 2%-6%.

Effective postclassification refinement method.

Enhanced consistency across temporal data sets.

Abstract

The current practice in land cover/land use change analysis relies heavily on the individually classified maps of the multitemporal data set. Due to varying acquisition conditions (e.g., illumination, sensors, seasonal differences), the classification maps yielded are often inconsistent through time for robust statistical analysis. 3D geometric features have been shown to be stable for assessing differences across the temporal data set. Therefore, in this article we investigate he use of a multitemporal orthophoto and digital surface model derived from satellite data for spatiotemporal classification. Our approach consists of two major steps: generating per-class probability distribution maps using the random-forest classifier with limited training samples, and making spatiotemporal inferences using an iterative 3D spatiotemporal filter operating on per-class probability maps. Our experimental results demonstrate that the proposed methods can consistently improve the individual classification results by 2%-6% and thus can be an important postclassification refinement approach.