Design And Implementation Of Real-time Overwriting Identification And Optimization Of Candidate

With the popularity of more and more touch screen mobile intelligent terminals,handwriting input method has drawn extensive attention. But owing to the limitation of the screen size, users can’t input handwriting continuously, which leads to the poor efficiency of input. In recent years, someone came up with the input mode,inputting continuous handwriting in the same area, that is to say, overlapped handwriting input, which can greatly improve the efficiency of handwriting input. At present, there are fashioned overlapped handwriting input methods, but their efficiency can’t be able to meet people’s needs, so there’s still room for great improvement.The main work of this paper is to realize a real-time overlapped handwriting input system which is convenient for user interaction. Through real-time calculating and updating the search path while users are inputting handwriting, plenty of timeconsuming operations can happen in parallel to the users’ handwriting input, in this way, results of recognition can be displayed promptly after input, all of the above can enhance the efficiency of handwriting input and enrich the user experience. There are several problems have been solved in the process of realizing real-time input task,including real-time over segmentation of handwriting, update of the mesh of segmentation to recognize candidate characters, real-time search of path,multithreaded operations and out-of-step refreshments between different interfaces.Considering that it is difficult to recognize all the handwriting for every input system,there are always errors for recognition, in order to introduce the significant factor,user interaction, this paper also developed the user-centered toolbar of division and merger. When users are unsatisfied to the results of segmentation of track, they can indicate and intervene the input system, then divide or merge the track and the candidate characters again.This system can meet the functional and nonfunctional requirements of users, it works well and the efficiency of handwriting input increases substantially. Thissystem is a project, other subsequent functions are being continued perfected, and it will be employed by more users in the future when it comes online.