Carnegie Airborne Observatory-2: Increasing science data dimensionality via high-fidelity multi-sensor fusion

详细信息    查看全文

• 作者：Gregory P. Asner ; David E. Knapp ; Joseph Boardman ; Robert O. Green ; Ty Kennedy-Bowdoin ; Michael Eastwood ; Roberta E. Martin ; Christopher Anderson ; Christopher B. Field
• 关键词：Airborne remote sensing ; CAO ; Data fusion ; Hyperspectral ; Imaging spectroscopy ; LiDAR ; Light detection and ranging ; Macroscale ecology
• 刊名：Remote Sensing of Environment
• 出版年：2012
• 出版时间：September, 2012
• 年：2012
• 卷：124
• 期：Complete
• 页码：454-465
• 全文大小：5293 K

文摘

The Carnegie Airborne Observatory (CAO) was developed to address a need for macroscale measurements that reveal the structural, functional and organismic composition of Earth's ecosystems. In 2011, we completed and launched the CAO-2 next generation Airborne Taxonomic Mapping Systems (AToMS), which includes a high-fidelity visible-to-shortwave infrared (VSWIR) imaging spectrometer (380-2510 nm), dual-laser waveform light detection and ranging (LiDAR) scanner, and high spatial resolution visible-to-near infrared (VNIR) imaging spectrometer (365-1052 nm). Here, we describe how multiple data streams from these sensors can be fused using hardware and software co-alignment and processing techniques. With these data streams, we quantitatively demonstrate that precision data fusion greatly increases the dimensionality of the ecological information derived from remote sensing. We compare the data dimensionality of two contrasting scenes ¡ª a built environment at Stanford University and a lowland tropical forest in Amazonia. Principal components analysis revealed 336 dimensions (degrees of freedom) in the Stanford case, and 218 dimensions in the Amazon. The Amazon case presents what could be the highest level of remotely sensed data dimensionality ever reported for a forested ecosystem. Simulated misalignment of data streams reduced the effective information content by up to 48 % , highlighting the critical role of achieving high precision when undertaking multi-sensor fusion. The instrumentation and methods described here are a pathfinder for future airborne applications undertaken by the National Ecological Observatory Network (NEON) and other organizations.