A hierarchical methodology framework for multisource data fusion in vegetation classification

This Letter presents a new methodological framework for a hierarchical data fusion system for vegetation classification using multi-sensor and multite mporal remotely sensed imagery. The uniqueness of the approach is that the overall structure of the fusion system is built upon a hierarchy of vegetat ion canopy attributes that can be remotely detected by sensors. The framewo rk consists of two key components: an automated multisource image registrat ion system and a hierarchical model for multi-sensor and multi-temporal dat a fusion.