A Hybrid Image Enhancement Algorithm for Effective Concrete Surface Crack Classification

Huge number of images are acquired and analysed every day for a range applications in civil infrastructure. One such application is the identification cracks concrete surface images, which challenge owing to their low contrast resolution, blurriness, noise information loss. Existing image enhancement algorithms improve either or resolution rather limited extent. This paper proposes Hybrid Image Enhancement (HIE) algorithm both using Wavelet transform Singular Value Decomposition (SVM). The enhanced crack classified into specific types. classification comprises preprocessing, detection, feature extraction classification. initially preprocessed Wiener filter remove following detected morphological operations discontinuities segmented regions eliminated K-Dimensional Tree algorithm. Features extracted from statistical geometric features. thereafter types three different neural network, kernel tree based categories. proposed HIE validated quantitative metrics results obtained compared with those State-of-the-Art methods datasets. have shown that offers significantly improved accuracy between 6% 10% images.