English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Crop classification via deep learning on ground imagery can deliver timelyand accurate crop-specific information to various stakeholders. Dedicatedground-based image acquisition exercises can help to collect data in datascarce regions, improve control on timing of collection, or when study areasare to small to monitor via satellite. Automatic labelling is essential whencollecting large volumes of data. One such data collection is the EU's Land UseCover Area frame Survey (LUCAS), and in particular, the recently publishedLUCAS Cover photos database. The aim of this paper is to select and publish asubset of LUCAS Cover photos for 12 mature major crops across the EU, todeploy, benchmark, and identify the best configuration of Mobile-net for theclassification task, to showcase the possibility of using entropy-based metricsfor post-processing of results, and finally to show the applications andlimitations of the model in a practical and policy relevant context. Inparticular, the usefulness of automatically identifying crops on geo-taggedphotos is illustrated in the context of the EU's Common Agricultural Policy.The work has produced a dataset of 169,460 images of mature crops for the 12classes, out of which 15,876 were manually selected as representing a cleansample without any foreign objects or unfavorable conditions. The bestperforming model achieved a Macro F1 (M-F1) of 0.75 on an imbalanced testdataset of 8,642 photos. Using metrics from information theory, namely - theEquivalence Reference Probability, resulted in achieving an increase of 6%. Themost unfavorable conditions for taking such images, across all crop classes,were found to be too early or late in the season. The proposed methodologyshows the possibility for using minimal auxiliary data, outside the imagesthemselves, in order to achieve a M-F1 of 0.817 for labelling between 12 majorEuropean crops.

Crop identification using deep learning on LUCAS crop cover photos