Vegetation Characterization and Classification With Multi-Source Remote Sensing Data

Dear Colleagues,

Geospatial technologies are developing more rapidly than ever before, causing Earth observation (EO) data to be widely available in unprecedented volume and detail. Extracting the useful information buried in all these data is a challenge that calls for innovative solutions. In parallel with this increase in data, there is now an increased demand for the accurate determination of a growing number of attributes of vegetation canopies using remote sensing for inventory and sustainable management; analysis of wild-life habitats and biodiversity; renewable energy production; precision agriculture and forestry; carbon stock estimation, etc. The accurate characterization of vegetation canopies, such as the determination of cover types and species and the retrieval of their biophysical parameters, remains challenging using traditional data and stand-alone approaches. Significant advancements in machine learning, deep learning, and artificial intelligence have wide implications for this field. Data/information fusion has lately been attracting much interest in the remote sensing community to generate maximal values from the available data sets. This topic collection, thus, aims to disseminate innovative research in vegetation characterization using multi-source remotely sensed data and accumulated domain knowledge. The papers may be related to, but not limited to, the following topics:

• Quantification and representation of uncertainty associated with information fusion.
• Integration of physical principles and domain knowledge to data-driven machine learning and deep learning methods.
• Development of novel information fusion algorithms for applications, such as: a. segmentation and classification of vegetation types; b. retrieval of physical/biophysical/biochemical parameters of vegetation canopies; c. change detection.
• Development of frameworks, methods, and tools for large-scale operational implementation in vegetation characterization using multi-source data.

Prof. Dr. Baoxin Hu
Prof. Dr. Linhai Jing
Topic Editors