Automatic crack segmentation using deep high-resolution representationlearning

Abstract

Cracks are one of the most common types of surface defects that occur on various engineering infrastructures. Visual-based crack detection is a challenging step due to the variation of size, shape, and appearance of cracks. Existing convolutional neural network (CNN)-based crack detection networks, typically using encoder-decoder architectures, may suffer from loss of spatial resolution in the high-to-low and low-to-high resolution processes, affecting the accuracy of prediction. Therefore, we propose <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:msup> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:mi mathvariant="normal">H</mml:mi> <mml:mi mathvariant="normal">R</mml:mi> <mml:mi mathvariant="normal">N</mml:mi> <mml:mi mathvariant="normal">e</mml:mi> <mml:mi mathvariant="normal">t</mml:mi> </mml:mrow> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:mi mathvariant="normal">e</mml:mi> </mml:mrow> </mml:msup> </mml:mrow> </mml:math> , an enhanced version of a high-resolution network (HRNet), by removing the downsampling operation in the initial stage, reducing the number of high-resolution representation layers, using dilated convolution, and introducing hierarchical feature integration. Experiments show that the proposed <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:msup> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:mi mathvariant="normal">H</mml:mi> <mml:mi mathvariant="normal">R</mml:mi> <mml:mi mathvariant="normal">N</mml:mi> <mml:mi mathvariant="normal">e</mml:mi> <mml:mi mathvariant="normal">t</mml:mi> </mml:mrow> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:mi mathvariant="normal">e</mml:mi> </mml:mrow> </mml:msup> </mml:mrow> </mml:math> with relatively few parameters can achieve more accuracy and robust performance than other recent approaches.

Topics & Concepts