A Survey of Thinning Techniques on Two Dimensional Binary Images

An image may be considered to contain sub-images sometimes referred to as regions-of-interest, ROIs, or simply regions. This concept reflects the fact that images frequently contain collections of objects. In a sophisticated image processing system it should be possible to apply specific image processing operations to selected regions. For performing image processing operations, thinning is much more essential and highly adaptive technique. Skeleton is the end product of the thinning technique. Skeleton is important shape descriptor in object representation and recognition and able to captures essential topology and shape information of the object in a simple form. So, thinning is extremely useful in solving various problems such as recognition, matching and retrieval in different types of image analysis. The aim of this paper is to survey the various thinning algorithms with their procedures, performance and limitations.