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/

Zendy Plus

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

Zendy Tools

Zendy Open

The vast network of bridges in the United States raises a high requirementfor maintenance and rehabilitation. The massive cost of manual visualinspection to assess bridge conditions is a burden to some extent. Advancedrobots have been leveraged to automate inspection data collection. Automatingthe segmentations of multiclass elements and surface defects on the elements inthe large volume of inspection image data would facilitate an efficient andeffective assessment of the bridge condition. Training separate single-tasknetworks for element parsing (i.e., semantic segmentation of multiclasselements) and defect segmentation fails to incorporate the close connectionbetween these two tasks. Both recognizable structural elements and apparentsurface defects are present in the inspection images. This paper is motivatedto develop a multitask deep learning model that fully utilizes suchinterdependence between bridge elements and defects to boost the model's taskperformance and generalization. Furthermore, the study investigated theeffectiveness of the proposed model designs for improving task performance,including feature decomposition, cross-talk sharing, and multi-objective lossfunction. A dataset with pixel-level labels of bridge elements and corrosionwas developed for model training and testing. Quantitative and qualitativeresults from evaluating the developed multitask deep model demonstrate itsadvantages over the single-task-based model not only in performance (2.59%higher mIoU on bridge parsing and 1.65% on corrosion segmentation) but also incomputational time and implementation capability.

A Multitask Deep Learning Model for Parsing Bridge Elements and Segmenting Defect in Bridge Inspection Images