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

Surface cracks on buildings, natural walls and underground mine tunnels canindicate serious structural integrity issues that threaten the safety of thestructure and people in the environment. Timely detection and monitoring ofcracks are crucial to managing these risks, especially if the systems can bemade highly automated through robots. Vision-based crack detection algorithmsusing deep neural networks have exhibited promise for structured surfaces suchas walls or civil engineering tunnels, but little work has addressed highlyunstructured environments such as rock cliffs and bare mining tunnels. Toaddress this challenge, this paper presents PointCrack3D, a new3D-point-cloud-based crack detection algorithm for unstructured surfaces. Themethod comprises three key components: an adaptive down-sampling method thatmaintains sufficient crack point density, a DNN that classifies each point ascrack or non-crack, and a post-processing clustering method that groups crackpoints into crack instances. The method was validated experimentally on a newlarge natural rock dataset, comprising coloured LIDAR point clouds spanningmore than 900 m^2 and 412 individual cracks. Results demonstrate a crackdetection rate of 97% overall and 100% for cracks with a maximum width of morethan 3 cm, significantly outperforming the state of the art. Furthermore, forcross-validation, PointCrack3D was applied to an entirely new dataset acquiredin different locations and not used at all in training and shown to detect 100%of its crack instances. We also characterise the relationship between detectionperformance, crack width and number of points per crack, providing a foundationupon which to make decisions about both practical deployments and futureresearch directions.

PointCrack3D: Crack Detection in Unstructured Environments using a 3D-Point-Cloud-Based Deep Neural Network