Abstract: This paper presents an on-line synchronization input method based on paper form and electronic one.It is designed to address the problem that while a paper form is filled out with the digital pen,the target location of corresponding electronic form is not accurate,and the recognition rate is not high.Furthermore,the users tend to switch so frequently between paper and display device in such circumstance that their interaction experience would be reduced dramatically.This method consists of error compensation on pen information and intelligent locating in the target cell with context and knowledge base.Based on statistical language model and dictionary,we design the recognition algorithm which can improve the recognizer and increase its recognition efficiency.Moreover,the results also will be added into the dictionary and the error correction module to improve the system's adaptability.The method proposed in this paper is based on on-line synchronization input frame-FSIF which combines paper form with electronic form during input process.It contains four parts: pen information correction converters,intelligent locator,ink splitter and adaptive handwriting recognizer.According to recognition and procedures,the handwriting information is carried out on collection,correction,change,locating,segmentation,classification and identification,etc.We verify the method in experiments.The results indicate that the method is able to improve the pen-based synchronization input on paper forms and electronic ones.