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

Precision weed management offers a promising solution for sustainablecropping systems through the use of chemical-reduced/non-chemical roboticweeding techniques, which apply suitable control tactics to individual weeds.Therefore, accurate identification of weed species plays a crucial role in suchsystems to enable precise, individualized weed treatment. This paper makes afirst comprehensive evaluation of deep transfer learning (DTL) for identifyingcommon weeds specific to cotton production systems in southern United States. Anew dataset for weed identification was created, consisting of 5187 colorimages of 15 weed classes collected under natural lighting conditions and atvaried weed growth stages, in cotton fields during the 2020 and 2021 fieldseasons. We evaluated 27 state-of-the-art deep learning models through transferlearning and established an extensive benchmark for the considered weedidentification task. DTL achieved high classification accuracy of F1 scoresexceeding 95%, requiring reasonably short training time (less than 2.5 hours)across models. ResNet101 achieved the best F1-score of 99.1% whereas 14 out ofthe 27 models achieved F1 scores exceeding 98.0%. However, the performance onminority weed classes with few training samples was less satisfactory formodels trained with a conventional, unweighted cross entropy loss function. Toaddress this issue, a weighted cross entropy loss function was adopted, whichachieved substantially improved accuracies for minority weed classes.Furthermore, a deep learning-based cosine similarity metrics was employed toanalyze the similarity among weed classes, assisting in the interpretation ofclassifications. Both the codes for model benchmarking and the weed dataset aremade publicly available, which expect to be be a valuable resource for futureresearch in weed identification and beyond.

Performance Evaluation of Deep Transfer Learning on Multiclass Identification of Common Weed Species in Cotton Production Systems