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

Aero-engine is the core component of aircraft and other spacecraft. Thehigh-speed rotating blades provide power by sucking in air and fullycombusting, and various defects will inevitably occur, threatening theoperation safety of aero-engine. Therefore, regular inspections are essentialfor such a complex system. However, existing traditional technology which isborescope inspection is labor-intensive, time-consuming, andexperience-dependent. To endow this technology with intelligence, a novelsuperpixel perception graph neural network (SPGNN) is proposed by utilizing amulti-stage graph convolutional network (MSGCN) for feature extraction andsuperpixel perception region proposal network (SPRPN) for region proposal.First, to capture complex and irregular textures, the images are transformedinto a series of patches, to obtain their graph representations. Then, MSGCNcomposed of several GCN blocks extracts graph structure features and performsgraph information processing at graph level. Last but not least, the SPRPN isproposed to generate perceptual bounding boxes by fusing graph representationfeatures and superpixel perception features. Therefore, the proposed SPGNNalways implements feature extraction and information transmission at the graphlevel in the whole SPGNN pipeline, and SPRPN and MSGNN mutually benefit fromeach other. To verify the effectiveness of SPGNN, we meticulously construct asimulated blade dataset with 3000 images. A public aluminum dataset is alsoused to validate the performances of different methods. The experimentalresults demonstrate that the proposed SPGNN has superior performance comparedwith the state-of-the-art methods. The source code will be available athttps://github.com/githbshang/SPGNN.

Superpixel Perception Graph Neural Network for Intelligent Defect Detection