Detection and Removal of Cracks and In Digitized Paintings

Detection and Removal of Cracks and In Digitized Paintings Project Abstract: There are many things which get breaks in the substrate like old paintings, the varnish or the paint. These breaks are called cracks or craquelure and caused due to drying, aging, and mechanical factors. Age cracks result because of nonuniform contraction in the canvas whereas drying cracks are due to the evaporation of components of volatile paint and shrinkage of the paint and mechanical cracks are caused with painting deformations like vibrations and impacts.

Hence digital image processing techniques are used to detect and remove the cracks on digitized paintings. It is also used as a nondestructive tool for the actual restoration. The user must select an area on each crack that to be restored. The formers generally based on partial differential equations (PDEs) and on the calculus of variations.

The technique includes the following stages. They are crack detection, separation of the thin dark brush strokes which are misidentified as cracks, and crack filling (interpolation).

The cracks are identified and detected by thresholding which is an integrated methodology for removal of cracks. Furthermore, a thin dark brush strokes are removed with the help of either a median radial basis function neural network or a semi-automatic procedure based on region growing. Ultimately, crack filling such as median filter is done.

The existing tools like PhotoShop has an option like blur. This option roughly pixel the transformations but there is no exact filling procedure in this existing system. The proposed system has exact crack detection and filling procedure which includes top-hat transformation, region-growing algorithm (grassfire algorithm) and median filter procedures.

Conclusion:

This paper is all about the cracks which are detected with the help of top-hat transform. The thin dark brush strokes are misidentified as cracks which are separated either by an automatic technique (MRBF networks) or by a semi-automatic approach.

Download Detection and Removal of Cracks and In Digitized Paintings ECE Final Year Project Report.