A fine extraction algorithm for image-based surface cracks in underwater dams

Abstract

Under the action of water erosion and self-aging, reservoir dams are prone to develop cracks, which affect safe operation. Underwater visual imaging can be used to detect dam surface cracks, but spalling, aquatic plants and suspended sediments result in low image contrast and complex backgrounds. With the use of unsupervised machine learning, this paper proposes a fine segmentation and extraction algorithm for image-based dam surface cracks. First, adaptive histogram equalization is used to change the uneven illumination areas of underwater surface images intoeven illumination areas, whose statistical characteristics are calculated under linear spatial filtering. Second, the extraction problem of crack areas of interest after dodging preprocessing is transformed into calculating the distance of the image block cluster center, which can distinguish the image blocks of crack features from the background interference features. Third, the fine extraction of crack images is carried out by considering the connected domains and morphological features, and the posterior probability of an image sample category is obtained based on the soft clustering of a Gaussian mixed model. Finally, different extraction algorithms related to surface cracks are evaluated in extensive experiments. The results validate the superior performance of the proposed extraction algorithm with 90.1% extraction accuracy, 6.5% missing alarm rate and 7.2% false alarm rate.