Study Of Handwriting Identification Methods

Writer identification (WI) is a technique that aims to decide the identity of writers according to the handwriting styles. In recent years, social backgrounds urge more achievements in computer WI, and advances in correlated disciplines such as pattern recognition and artificial intelligence supply chances for the development of computer WI. Under this background, the application background and history of development of WI techniques are briefly introduced in this thesis. The state of the WI researches is also surveyed. The nature of the WI problem and difficulties are analyzed, and then the overall strategy and the plan of building computer WI system are proposed. It is pointed out that an applicable computer WI system is human-computer interactive and final decision by human experts. In respect to the preprocessing of document image, a general noise elimination approach and an effective character normalization approach are presented. In the process of feature character comparison, some methods for character comparison are presented which are chamfer matching, the texture feature of gray matrix and texture analysis using Wigner distribution. These methods are motivated by advanced pattern recognition and computer vision methodology, and have achieved promising results in writer recognition and verification experiments. In this thesis, the method of similar estimation (the nearest neighbor) is applied to classify the feature character. According to different methods of feather extraction, distance average and swatch average are used respectively in the process of distance estimation, which furthermore reflect the whole information of lots of swatch belonging to the same type. The experiment has proven that the techniques of handwriting feather extraction and decision can get the satisfying right identifying rate.