.png){kind=icon}
Return
Image-processing-based automatic crack detection and classification for refractory evaluation
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ceramint.2022.04.307.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Crack generation and propagation in refractories is inevitable during their service life, which ultimately contributes to their failure. Accordingly, this study proposes a method that includes a watershed algorithm and classification criterion to quantitatively evaluate the fracture behavior in refractories. Additionally, to demonstrate the stability and reliability of the algorithm, its crack analysis results were compared with those of manual methods for alumina-magnesia ramming materials and corundum castables. The results indicated that the algorithm was capable of automatically identifying, extracting, and classifying cracks. Additionally, the type of crack propagation and its length and width were analyzed to determine the performance and properties of refractories. The proposed method was highly effective and accurate for the automated evaluation of material fracture behavior.
Keywords:
Refractory
Classification criterion
Crack statistics
Watershed algorithm
Image processing
Journal
IF:
5.6
Papers:
5.1W
Citations:
15.5W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveEnhanced performance of low-carbon MgO-C refractories with nano-sized ZrO2-Al2O3 composite powder
PREAI
ShareSave
ShareSaveShareSaveShareSave
View more