<![CDATA[The complexity of multimedia contents is significantly increasing in the current world. This leads to an exigent demand for developing highly effective systems tosatisfy human needs. Until today, handwritten signature considered an important means that is used in banks and businesses to evidence identity, so there are many works triedto develop a method for recognition purpose. This paper introduced an efficient technique for offline signature recognition depending on extracting the local feature by utilizing the haar wavelet subbands and energy. Three different setsof features are utilized by partitioning the signature image into non overlapping blocks where different block sizes are used. CEDAR signature database is used asa dataset for testing purpose. The results achieved by this technique indicate a high performance in signature recognition.]]>
