FoSA: F* Seed-growing Approach for crack-line detection from pavement images

An Improved Neuron Segmentation Model for Crack Detection – Image Segmentation Model

“…Scable local percolation-based image processing techniques are tried in [9] to get the efficient and accurate detection result even for large surface images [10]. Some other local methods such as statistical filter [11], morphological filter [12], Hessian matrix-based filter [13] and wavelets [14,15] are also applied successfully when the texture of surfaces is fine grained; the performance will deteriorate significantly and sharply [16,17] when surface is coarse-grained and cracks are not continuous. The classical Principal Component Analysis (PCA) based algorithm [18] is adopted to detect cracks on a bridge surface, in which case the accuracy of results varies with camera pose and capturing distance where images are taken.…”

“…Among them, the minimal path-based methods [16,[21][22][23][24][25], which appear to be most prevalent in recent literature, consider a crack as a path of minimal length in a certain metric and then convert the crack detection problem to that of finding the minimal path. Similarly, the cracks with complicated geometrical changes such as branches and curves could also be coped in [26][27][28].…”

“…Fast and accurate detection is possible using the above mentioned region growing approaches if the crack structure is morphologically simple and the image noise level is low. It is extremely difficult to cope with the complicated topology changes of cracks such as disconnections and closed substructures, and only [16,23] are able to exract cracks with arbitrary topology.…”

Extraction of microcracks in rock images based on heuristic graph searching and application

“…For some poor condition images, Dare et al (2002) delineated the crack in a concrete structure as a polyline and measured the crack width along this polyline. Li et al (2011) andZou et al (2012) detected crack lines on pavement images with grainlike characteristics.…”

Climber-Inspired Fuzzy Logic Approach to Crack Localization Using Image Analysis

“…The digital images taken during the performed experiment or investigation represent the input data. The use of digital image processing and analysis methods for the purpose of localization of cracks is considered to be a common procedure (Adhikari et al, 2014;Li et al, 2013Li et al, , 2011Valença et al, 2013). The conventional techniques, such as thresholding and subsequent mathematical morphology procedures without any previous supporting classification, often lead to development of complex algorithms in order to deal with various difficulties such as noise, formation of parasitic spurs, non-uniform illumination, and poor resolution (Wong et al, 2006;Yazid and Arof, 2013).…”