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

Deep Convolutional Neural Networks (CNNs), such as Dense ConvolutionalNetworks (DenseNet), have achieved great success for image representation bydiscovering deep hierarchical information. However, most existing networkssimply stacks the convolutional layers and hence failing to fully discoverlocal and global feature information among layers. In this paper, we mainlyexplore how to enhance the local and global dense feature flow by exploitinghierarchical features fully from all the convolution layers. Technically, wepropose an efficient and effective CNN framework, i.e., Fast Dense ResidualNetwork (FDRN), for text recognition. To construct FDRN, we propose a new fastresidual dense block (f-RDB) to retain the ability of local feature fusion andlocal residual learning of original RDB, which can reduce the computing effortsat the same time. After fully learning local residual dense features, weutilize the sum operation and several f-RDBs to define a new block termedglobal dense block (GDB) by imitating the construction of dense blocks to learnglobal dense residual features adaptively in a holistic way. Finally, we usetwo convolution layers to construct a down-sampling block to reduce the globalfeature size and extract deeper features. Extensive simulations show that FDRNobtains the enhanced recognition results, compared with other related models.

Dense Residual Network: Enhancing Global Dense Feature Flow for Character Recognition