Abstract:OCR is a complicated process, there are many factors that can influence the recognition rate.Early period people tried to optimize the classifier to obtain high recognition rate, but the premise is that there is only one character no matter print or handwritten.For the performance of classifier has been promoted a lot, recognition rate for single character is high enough for commercial use.With the development of the demand for handwritten text recognition, how to raise the recognition rate of OCR system becomes very important.Unlike OCR system for print which focus on classifier.The research of OCR system for handwritten text is mainly on character segmentation.Statistical analysis showed that the mistake made by missegment is more than the mistake made by classifier.This is decided by the feature of handwritten text.There are more randomness and the lines are not horizontal, besides that, handwritten Chinese characters are more like overlapped and the gaps between characters are smaller.So this is the difficulty of handwritten Chinese characters.In this paper, the mutil-step searching nonlinear line exaction algorithm the paper proposed is easy and the accuracy is high, which can tackle the some weaknesses of direct projection method and indirect projection.

A Method of Handwriting Texts and Shapes Separation

Space Handwriting Character Recognition Method Based on Elimination of Connected Stroke

Improvements on EMG-based handwriting recognition with DTW algorithm.

Separation Method for Handwritten Shape and Text Based on Improved Stroke Entropy

Touching Character Separation in Chinese Handwriting Using Visibility-Based Foreground Analysis

Adaboost Based Handwritten/Printed Discrimination On Single Character

Improving Chinese Writer Identification by Fusion of Text-dependent and Text-independent Methods

Biometric personal identification based on handwriting

Enhanced Handwritten Character Recognition Algorithm Based on Template Matching Method

Off-Line Chinese Handwriting Identification Based on Stroke Shape and Structure

Handwritten Character Recognition Research Based on Adaptive Minimum Distance Classifiers Integration

AdaBoost-based handwritten/printed discrimination on a single character

A Chinese Handwriting Education System with Automatic Error Detection

Text-Independent writer identification based on fusion of dynamic and static features

Chinese Handwriting Identification Method Based on Keyword Extraction.

Combining Multiple Classifiers Based on Statistical Method for Handwritten Chinese Character Recognition

Robust End-to-End Offline Chinese Handwriting Text Page Spotter with Text Kernel

Pre-merger Method of Chinese Online Handwritten Text Based on Classification

Bag of Features Approach for Offline Text-Independent Chinese Writer Identification

Research of Chinese Handwritten Text Segmentation Algorithm

Chinese handwriting quality evaluation based on analysis of recognition confidence