A research team from Brazil are using Caffe and GeForce to identify crops from remote sensing images which is essential to know and monitor the land-use, or to estimate the feasible production amount. Identifying crops from remote sensing images is fundamental to know and monitor land-use. However, manual identification is expensive and maybe impracticable given the amount data.
Automatic methods, although interesting, are highly dependent on the quality of extracted features, since encoding the spatial features in an efficient and robust fashion is the key to generating discriminatory models.
Even though many visual descriptors have been proposed or successfully used to encode spatial features, in some cases, more specific descriptions are needed. Deep learning has achieved very good results in some tasks, mainly boosted by the feature learning performed which allows the method to extract specific and adaptable visual features depending on the data.
A new research paper proposes two multi-scale methods, based on deep learning, to identify coffee crops. The team conducted a systematic evaluation of the proposed algorithms using a remote sensing dataset.