Road Crack Detection using Rooted Ratio-Dependent Scaling Factor and Pixel Difference based Fuzzy-C Means Clustering Technique

One of the biggest issues facing the world today is road cracks. It might lead to increased road safety and a higher likelihood of accidents. An enormous amount of funding is set aside year for road upkeep and repairs. Early detection of the cracks can lower this expense. The majority of labourers are needed if this detection is done manually. Manual detection will take longer and provide less precision. To determine the precise region of the crack, numerous techniques have been devised in the past. In owing of background noise, sparse contrast, & intensity dissimilarity, fracture detection from photographs is a complex process. This study presents a new approach called PED-FCMC, introduces a pixel enhancement technique using a rooted ratio-dependent scaling factor combined with pixel differences from the total pixel sum, integrated into the Fuzzy C- Means clustering algorithm. The PED-FCMC notably enhances fracture detection accuracy, allowing successful identification even in low-contrast images. Additionally, the algorithm is computationally efficient, requiring no extensive parameter adjustments. Experimental results demonstrate that this new segmentation technique outperforms both Fuzzy C-Means and K-means clustering in detecting various cracks in images, as measured by the Xie-Beni Index, Davies-Bouldin Index, Partition Index, and overall execution time.