Watermarking Based on the Classification of Cracks in Paintings

This system presents a methodology for watermarking by detection of cracks in images of old statues of temples and old paintings. The cracks are detected by thresholding the output of the morphological top-hat transform. Afterwards, the thin dark brush strokes, which have been misidentified as cracks are removed using a semi-automatic procedure based on region growing. The cracks are classified using an unsupervised approach, which incorporates fuzzy clustering of the patterns. This system classifies cracks using the fuzzy k-means technique. The watermarked information is stored on the specific category of cracks. The user gives the input of the category of crack. The resulting image is the watermarked image. While retrieving the watermark information, the above procedure for identifying cracks is conducted and the watermark information is retrieved from the same category of cracks specified during the hiding stage.