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

Single-pixel imaging (SPI) is a novel, unconventional method that goes beyondthe notion of traditional cameras but can be computationally expensive and slowfor real-time applications. Deep learning has been proposed as an alternativeapproach for solving the SPI reconstruction problem, but a detailed analysis ofits performance and generated basis patterns when used for SPI is limited. Wepresent a modified deep convolutional autoencoder network (DCAN) for SPI on64x64 pixel images with up to 6.25% compression ratio and apply binary andorthogonality regularizers during training. Training a DCAN with theseregularizers allows it to learn multiple measurement bases that havecombinations of binary or non-binary, and orthogonal or non-orthogonalpatterns. We compare the reconstruction quality, orthogonality of the patterns,and robustness to noise of the resulting DCAN models to traditional SPIreconstruction algorithms (such as Total Variation minimization and FourierTransform). Our DCAN models can be trained to be robust to noise while stillhaving fast enough reconstruction times (~3 ms per frame) to be viable forreal-time imaging.

Deep-learned orthogonal basis patterns for fast, noise-robust single-pixel imaging