Modified component-labeling algorithms applied to grayscale images

In this paper, we proposed connected-component labeling algorithms applied to grayscale images. To develop the algorithms, three representative labeling algorithms for binary images which are raster-scan & label-equivalence-solving algorithm, searching & label propagation algorithms, and contour-tracing & label propagation algorithm are modified. The performance of these algorithms is evaluated in point of view of computational time consumption. Through the experiments, the advantages and disadvantages of each algorithm are compared. The modified algorithms and its results of test will be useful in labeling of various grayscale images and their applications.