Dissertation > Excellent graduate degree dissertation topics show

On-Line Handwritten Chinese Character Recognition Approach Based on Sentence Level

Author: GuoXinZuo
Tutor: ChenQingCai
School: Harbin Institute of Technology
Course: Computer Science and Technology
Keywords: On-line handwriting recognition Statement-level Feature Extraction AP Clustering Language model
CLC: TP391.43
Type: Master's thesis
Year: 2010
Downloads: 54
Quote: 0
Read: Download Dissertation

Abstract


Online handwritten Chinese character input technology is already very mature. Plays a key role in the online recognition algorithm can not achieve a high accuracy level. The main reason is the category of handwritten Chinese characters, shaped diversified forms nearly word, even the pen input and other reasons. Therefore, the characteristics and classification recognition technology is the key to how to resolve these differences, and to elect the most classification ability. The mature high recognition rate of handwriting recognition products are constantly updated to come out, but they are limited to an input area and only identify a character at a time, but in fact there is a sentence or paragraph context semantics. Select the correct target kanji word input, often from more than one candidate word, interrupting the original writing ideas. Tablet PC, a large screen touch device like traditional paper input possibilities, one can enter Chinese characters or a statement. So, a \Line handwritten Chinese character recognition technology will also pay more attention to the statement as well as the chapter level of direction. In this paper, a statement-level on-line handwriting recognition system will be based on a statement-level recognition algorithm. Users can under conditions of unconstrained handwriting recognition process in the background. When the user completes a paragraph, you can one-time the whole text to identify and displayed and saved together with the handwritten document. The whole process is divided into stages of word recognition stage and post-processing language model. Word recognition accuracy is an important factor, especially for the feature extraction. The continuous NCFE eight directions features extraction algorithm to improve the endpoint of the vector processing, the use of different methods of assignment according to different situations, and enhance the characteristics of the distinction between performance, while AP clustering algorithm is introduced to identify the rough classification stage, AP clustering algorithm and other clustering algorithms have a distinct advantage compared to handwritten character recognition. The second stage takes full advantage of the semantic information entered by the user to create a language model based on a statement-level candidate recognition results to adjust the entire paragraph. Here the capacity of the dictionary and fields a great influence on the result of the adjustment, simply dependent dictionary adjustment results are sometimes not only did not improve the recognition rate, but affect to the correct recognition result. To solve this problem, the establishment of the word recognition result with the language model by combining optimized weight recognition algorithm. Finally, the experiment testing algorithm in the HIT-OR3C as well as the Chinese Academy of Sciences CASIA-OLHWDB1 handwritten data set. The results showed that compared with other methods, the accuracy rate has been significantly improved. Entire statement-level recognition system with the existing recognition system has ease of use. Recognition accuracy can meet the requirements.

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. ISAR Imaging Simulation of Space Targets and Target Recognition Based on ISAR Images,TN957.52
  3. Research on Structure Transition Technology for SMT,TP391.2
  4. Research on Feature Extraction and Classification of Pulse Waveform for Cholecystitis and Nephrotic Syndrome Diagnosis,TP391.41
  5. Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
  6. Word Sense Disambiguation Corpus Automatic Acquisition,TP391.1
  7. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  8. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  9. Research on Visual Measurement for Spacecraft Rendezvous and Approach,TP391.41
  10. Research on the Image Real-Time Acquisition, Storage and Image Processing System,TP391.41
  11. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  12. Multi-currency Notes Technology Research and Implementation,TP391.41
  13. The Research on Paper Currency Classification Method Based on Harr-Like Feature and Minimal Ball Including Samples,TP391.41
  14. Pavement Distress Recognition Based on Image,TP391.41
  15. Research on Visual Detection and Tracking of Mobile Robots,TP242.62
  16. Research on Fusion Algorithm of Hyper Spectral and High Spatial Resolution Remote Sensing Image,TP751
  17. An Approach for Identifying a Plant Resistance Gene Based on the Random Forest,Q943
  18. Tobacco Diseases Auto-Recognition Research Based on Image Processing Technology,S435.72
  19. Research on Nondestructive Detection Technology for External Qualities of Papayas Based-on Vision,S667.9
  20. Research on Identification System of Cashmere and Wool Fiber,TS101.921
  21. Research for Infrared Image Target Identification and Tracking Technology,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Character recognition devices
© 2012 www.DissertationTopic.Net  Mobile