The Feature Extraction Research Of Handwritten Uyghur Words Based On Android Mobile Terminals

Uyghur information processing technologies play an important role in the Xinjiang Uygur Autonomous Region. However, the existing Uyghur on-line handwritten recognition technologies cannot meet the need of the handwritten Uyghur words input problem. In this paper, we take the handwritten Uyghur words features which are comprised of the word-part features as the research object, make the word as the recognition unit. Combined the writing Uyghur characteristics, the characteristics of the mobile terminal and the graph and image technology performed the framework of the handwritten Uyghur words recognition based on the Android mobile terminals and the methods of the handwritten Uyghur words feature extraction. Try to solve the difficulties of the character recognition caused by the cursive characteristic of the Uyghur and the basic letter boundary is hard to identify;and try to make the tentative exploration of the handwritten Uyghur words recognition in the mobile terminals.1) In order to solve the collection problem of handwritten word samples to research handwritten Uyghur words input and data collection methods based on the Android mobile device.2) Find out the acquisition approach of the handwritten Uyghur word-parts and the affixes. 3) Research and analyses the different feature selection and extraction of the word-part, to build handwritten Uyghur words characteristic sequence as the handwritten Uyghur recognition unit.This paper introduces the prepared work for handwritten Uyghur recognition based on the Android mobile terminals, and it has provided help and reference for the further study of the handwritten Uyghur words analysis and recognition research based on the Android mobile terminals.