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

Data-driven deep learning methods have shown great potential in croplandmapping. However, due to multiple factors such as attributes of cropland(topography, climate, crop type) and imaging conditions (viewing angle,illumination, scale), croplands under different scenes demonstrate a greatdomain gap. This makes it difficult for models trained in the specific scenesto directly generalize to other scenes. A common way to handle this problem isthrough the "Pretrain+Fine-tuning" paradigm. Unfortunately, considering thevariety of features of cropland that are affected by multiple factors, it ishardly to handle the complex domain gap between pre-trained data and targetdata using only sparse fine-tuned samples as general constraints. Moreover, asthe number of model parameters grows, fine-tuning is no longer an easy andlow-cost task. With the emergence of prompt learning via visual foundationmodels, the "Pretrain+Prompting" paradigm redesigns the optimization target byintroducing individual prompts for each single sample. This simplifies thedomain adaption from generic to specific scenes during model reasoningprocesses. Therefore, we introduce the "Pretrain+Prompting" paradigm tointerpreting cropland scenes and design the auto-prompting (APT) method basedon freely available global land cover product. It can achieve a fine-grainedadaptation process from generic scenes to specialized cropland scenes withoutintroducing additional label costs. To our best knowledge, this work pioneersthe exploration of the domain adaption problems for cropland mapping underprompt learning perspectives. Our experiments using two sub-meter croplanddatasets from southern and northern China demonstrated that the proposed methodvia visual foundation models outperforms traditional supervised learning andfine-tuning approaches in the field of remote sensing.

Using Global Land Cover Product as Prompt for Cropland Mapping via Visual Foundation Model