ABSTRACT
In order to solve the problems that the excessive parameters of the network model in the existing handwritten character recognition methods increase the consumption of computing and storage resources, the target recognition rate is difficult to meet the real-time requirements, and it is easy to cause parameter explosion, gradient loss and over-fitting. Based on the traditional digital image processing module, a long-term and short-term memory network (LSTM) model is embedded to identify the handwritten score template of a specific test paper. This method performs normalization, noise reduction, tilt correction, and binarization preprocessing on the handwritten digital test paper image. The Hough line detection is used to cut the horizontal line, and the LSD algorithm erases the vertical line to obtain the target image. Finally, a new digital recognition algorithm is constructed based on the LSTM model. Through experimental testing and verification, the results show that the method effectively improves the rate of recognizing handwritten digit scores in pictures, reduces the number of model parameters and resource consumption, accelerates the convergence speed of the model, and meets the actual use requirements of identifying specific test paper picture templates under certain conditions.
- Zhang Huamei, Zhang Jiaojie. Review of off-line handwritten digit recognition based on artificial intelligence [J]. Journal of Nanjing University of Posts and Telecommunications (Natural Science Edition), 2021, 41 ( 05 ) : 83-91.Google Scholar
- Wang Mei, Li Dongxu. Handwritten Numeral Recognition Based on Improved VGG-16 and Naive Bayes [J]. Modern Electronic Technology, 2020, 43 (12) : 176-181+186.Google Scholar
- LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J].Proceedings of theIEEE,1998,86( 11) ; 2278-2324.Google Scholar
- Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems, pages 1097–1105, 2012.Google ScholarDigital Library
- Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich. Going deeper with convolutions. In Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, pages 1-9, 2015.Google ScholarCross Ref
- Ahlawat S, Rishi R. Off-line handwritten numeral recognition using hybrid feature set-A comparative analysis[J]. Procedia Computer Science, 2017, 122: 1 092-1 099.Google ScholarCross Ref
- Srivastava S, Yadav S, Agrawalla K, Recognition of handwritten digits using computer vision preprocessor based combined architecture of self-organizing map and backpropagation on MNIST dataset[C]//2018 International Conference on Recent Innovations in Electrical, Electronics & Communication Engineering (ICRIEECE). 2018.Google Scholar
- Yu Shengxin, Xia Chengqi, Tang Zetian, Ding Zhao, Yang Chen. Handwritten digit recognition based on improved Inception convolutional neural network [J]. Computer application and software, 2019,36 (12) : 143-149.Google Scholar
- Zhuang Wei, Lei Xiaofeng, Song Fengtai, Dai Bin, Xie Kunqing. A Handwritten Numeral Recognition Method Based on Rotational Projection Statistical Features [J]. Computer Science, 2011,38 (11) : 278-281 + 302.Google Scholar
- Application and research of Zhang Yanfang.Bayesian network in handwritten digit recognition [D].North China Electric Power University, 2011.Google Scholar
- Li Boyan, Zhang Yong, Yuan Derong, Xiong Tangtang, He Lang. Handwritten digit recognition based on attention mechanism [J]. Computer Science, 2022,49 (S2): 626-630.Google Scholar
- Li Xueyu, Yang Song, Qin Weiming, Sun Shuting, Gong Ming, Qu Zhou. Research on license plate recognition algorithm based on Shumeipai and OpenCV [J].Computer programming skills and maintenance, 2019 ( 04 ) : 146-148.Google Scholar
- Alexander K. Seewald. AUTOMATIC EXTRACTION OF GO GAME POSITIONS FROM IMAGES: A MULTI-STRATEGICAL APPROACH TO CONSTRAINED MULTI-OBJECT RECOGNITION[J]. Applied Artificial Intelligence,2010,24(3):233-252.Google Scholar
- HOCHREITER S,SCHMIDHUBER J.Long Short-Term Memory[J].Neural computation, 1997,9(8):1735-1780.Google Scholar
Index Terms
- Research on Handwritten Test Paper Score Recognition Algorithm based on LSTM model
Recommendations
Semantic-Based Handwritten Chinese Character Recognition Model
ICCMS '10: Proceedings of the 2010 Second International Conference on Computer Modeling and Simulation - Volume 01There have been many different literals discussing algorithms for handwritten Chinese character recognition, but most algorithms aim at recognizing isolated Chinese character one by one. Therefore, their recognition accuracy isn’t good enough for the ...
Handwritten Cursive Jawi Character Recognition: A Survey
CGIV '08: Proceedings of the 2008 Fifth International Conference on Computer Graphics, Imaging and VisualisationThe subject of cursive handwritten character recognition is still open to be studied because of its complex nature. Recognition of Arabic handwritten and its variants such as Farsi (Persian) and Urdu had been receiving considerable attention in recent ...
Handwritten ZIP code recognition using lexicon free word recognition algorithm
ICDAR '95: Proceedings of the Third International Conference on Document Analysis and Recognition (Volume 2) - Volume 2The paper describes a new approach to ZIP code recognition using a word recognition algorithm, where a numeral string is recognized as a word. The paper also describes an end to end ZIP code recognition system consisting of tilt/slant correction, line ...
Comments