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

Recently, social infrastructure is aging, and its predictive maintenance hasbecome important issue. To monitor the state of infrastructures, bridgeinspection is performed by human eye or bay drone. For diagnosis, primarydamage region are recognized for repair targets. But, the degradation at worselevel has rarely occurred, and the damage regions of interest are often narrow,so their ratio per image is extremely small pixel count, as experienced 0.6 to1.5 percent. The both scarcity and imbalance property on the damage region ofinterest influences limited performance to detect damage. If additional dataset of damaged images can be generated, it may enable to improve accuracy indamage region segmentation algorithm. We propose a synthetic augmentationprocedure to generate damaged images using the image-to-image translationmapping from the tri-categorical label that consists the both semantic labeland structure edge to the real damage image. We use the Sobel gradient operatorto enhance structure edge. Actually, in case of bridge inspection, we apply theRC concrete structure with the number of 208 eye-inspection photos that rebarexposure have occurred, which are prepared 840 block images with size 224 by224. We applied popular per-pixel segmentation algorithms such as the FCN-8s,SegNet, and DeepLabv3+Xception-v2. We demonstrates that re-training a data setadded with synthetic augmentation procedure make higher accuracy based onindices the mean IoU, damage region of interest IoU, precision, recall, BFscore when we predict test images.

Synthetic Image Augmentation for Damage Region Segmentation using Conditional GAN with Structure Edge