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 recent advancement in deep Convolutional Neural Network (CNN) has broughtinsight into the automation of X-ray security screening for aviation securityand beyond. Here, we explore the viability of two recent end-to-end objectdetection CNN architectures, Cascade R-CNN and FreeAnchor, for prohibited itemdetection by balancing processing time and the impact of image data compressionfrom an operational viewpoint. Overall, we achieve maximal detectionperformance using a FreeAnchor architecture with a ResNet50 backbone, obtainingmean Average Precision (mAP) of 87.7 and 85.8 for using the OPIXray and SIXraybenchmark datasets, showing superior performance over prior work on both. Withfewer parameters and less training time, FreeAnchor achieves the highestdetection inference speed of ~13 fps (3.9 ms per image). Furthermore, weevaluate the impact of lossy image compression upon detector performance. TheCNN models display substantial resilience to the lossy compression, resultingin only a 1.1% decrease in mAP at the JPEG compression level of 50.Additionally, a thorough evaluation of data augmentation techniques isprovided, including adaptions of MixUp and CutMix strategy as well as otherstandard transformations, further improving the detection accuracy.

Operationalizing Convolutional Neural Network Architectures for Prohibited Object Detection in X-Ray Imagery