Land Usage Classification: A Hierarchical Neural Network Approach
The classification of land usage in grassland mountain bovine areas is important for the management of forage production and grazing in grass-based livestock systems. This article proposes a novel, hierarchical neural network based, approach towards the classification of land usage in these areas. A survey of 72 farms was conducted in the Massif Central (France). Information was gathered on geographical characteristics and cutting and/or grazing practices on three general groups of fields: cut only, cut and grazed, and grazed only fields. To classify the land usage, the data was clustered and visualised in a hierarchical fashion. This was done by using a novel method for the analysis and classification of data based on growing hierarchical self- organising maps (GHSOM). Self-organising maps have shown to be successful for the analysis of highly dimensional input data in data mining applications as well as for data visualisation. Moreover, the hierarchical architecture of the GHSOM is more flexible than a single SOM in the adaptation process to input data, capturing inherent hierarchical relationships among them. Experimental results show the utility of this approach.
Citation : Palomo, E.J., Elizondo, D. and Brunschwig, G. (2014) Land Usage Classification: A Hierarchical Neural Network Approach. Journal of Agricultural Science, 152 (05), pp. 817-828
Research Group : DIGITS
Research Institute : Institute of Artificial Intelligence (IAI)
Peer Reviewed : Yes