Offline Hand Writer Identification Based on Scale Invariant Feature Transform

The identification of a person on the basis of scanned images of handwriting is a useful biometric modality with application in forensic and historic document analysis. An efficient method for text-independent writer identification using a codebook method is introduced. The method uses the occurrence histogram of the shapes in a codebook to create a feature vector for each specific manuscript. Offline text-independent writer identification method based on scale invariant feature transform (SIFT). It include three stages training, enrollment, and identification stages. In three stages, an isotropic LoG filter is first used to segment the handwriting image into word regions (WRs). Then, the SIFT descriptors of word region and the corresponding scales and orientations (SOs) are extracted. In the second stage, an SD codebook is constructed by clustering the SDs of training samples. In the third stage, the SDs of the input handwriting are adopted to form an SD signature (SDS) by looking up the SD codebook and the SOs are utilized to generate a scale and orientation histogram (SOH). In the identification stage, the SDS and SOH of the input handwriting are extracted and matched with the enrolled ones for identification. Here we extracted six public data set. We also proposed a method that is k-means clustering instead of neural network which provides more efficiency.