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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Micropaleontology in geosciences focuses on studying the evolution ofmicrofossils (e.g., foraminifera) through geological records to reconstructpast environmental and climatic conditions. This field heavily relies on visualrecognition of microfossil features, making it suitable for computer visiontechnology, specifically deep convolutional neural networks (CNNs), to automateand optimize microfossil identification and classification. However, theapplication of deep learning in micropaleontology is hindered by limitedavailability of high-quality, high-resolution labeled fossil images and thesignificant manual labeling effort required by experts. To address thesechallenges, we propose a novel deep learning workflow combining hierarchicalvision transformers with style-based generative adversarial network algorithmsto efficiently acquire and synthetically generate realistic high-resolutionlabeled datasets of micropaleontology in large volumes. Our study shows thatthis workflow can generate high-resolution images with a high signal-to-noiseratio (39.1 dB) and realistic synthetic images with a Frechet inceptiondistance similarity score of 14.88. Additionally, our workflow provides a largevolume of self-labeled datasets for model benchmarking and various downstreamvisual tasks, including fossil classification and segmentation. For the firsttime, we performed few-shot semantic segmentation of different foraminiferachambers on both generated and synthetic images with high accuracy. This novelmeta-learning approach is only possible with the availability ofhigh-resolution, high-volume labeled datasets. Our deep learning-based workflowshows promise in advancing and optimizing micropaleontological research andother visual-dependent geological analyses.

ForamViT-GAN: Exploring New Paradigms in Deep Learning for Micropaleontological Image Analysis