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

The manufacturing of light-emitting diodes is a complexsemiconductor-manufacturing process, interspersed with different measurements.Among the employed measurements, photoluminescence imaging has severaladvantages, namely being a non-destructive, fast and thus cost-effectivemeasurement. On a photoluminescence measurement image of an LED wafer, everypixel corresponds to an LED chip's brightness after photo-excitation, revealingchip performance information. However, generating a chip-fine defect map of theLED wafer, based on photoluminescence images, proves challenging for multiplereasons: on the one hand, the measured brightness values vary from image toimage, in addition to local spots of differing brightness. On the other hand,certain defect structures may assume multiple shapes, sizes and brightnessgradients, where salient brightness values may correspond to defective LEDchips, measurement artefacts or non-defective structures. In this work, werevisit the creation of chip-fine defect maps using fully convolutionalnetworks and show that the problem of segmenting objects at multiple scales canbe improved by the incorporation of densely connected convolutional blocks andatrous spatial pyramid pooling modules. We also share implementation detailsand our experiences with training networks with small datasets of measurementimages. The proposed architecture significantly improves the segmentationaccuracy of highly variable defect structures over our previous version.

Rethinking Fully Convolutional Networks for the Analysis of Photoluminescence Wafer Images