A multi-scale canny detection technique for automatic assessment of fatigue cracking using morphological operators

The maintenance of existing pavements with a good quality is one of the key challenges to the pavement engineers. The pavement condition monitoring and evaluation is a complex task to proceed for making decisions regarding the appropriate maintenance strategies. Generally, the Pavement Condition Evaluation involves in measuring the roughness, skidding index and pavement distresses. The fatigue cracking is one of the major distresses occurring in flexible pavements. The present research study focused on detecting the fatigue cracking using the Digital Image Processing (DIP). The image preprocessing techniques like linear-smoothing technique and bilateral filtering were used for removing the noise and background non-uniform illumination. Later, Canny’s detection technique and morphological operators were used for enhancing the detection-accuracy. Finally, the resultant output was obtained via python programming and concluded that the proposed algorithm showed better results compared to the techniques without morphological operators.