|
|
基于PSO改进的BP神经网络字符识别算法研究
|
Abstract:
针对字符识别算法中常规的BP神经网络算法在训练识别多种类字符时易造成过拟合而导致训练失败，字符识别率不高，传统的模板匹配法对字符识别的效果也不好的问题，提出一种基于PSO改进的BP神经网络字符识别算法。采用粒子群优化算法(PSO)计算BP神经网络的初始权值和偏置，提高了网络的收敛速度和训练过程的稳定性；在常规BP神经网络的基础上加入正则化算法，防止网络训练的过拟合。在选择网络层节点参数的细节上，简化输出层节点数，减少网络的训练参数。最后通过对印刷体字符样本进行识别实验，识别准确率达到98.97%，证明该方法可有效地提高字符的识别率。
In order to solve the problem that the conventional BP neural network algorithm in character recognition algorithm is easy to cause over-fitting and leads to training failure, the character recognition rate is not high, and the effect of traditional template matching method on character recognition is not good, an improved BP neural network character recognition algorithm based on PSO is proposed. Particle swarm optimization algorithm (PSO) is used to calculate the initial weight and offset of BP neural network, which improves the convergence speed of network and the stability of training process. Regularization algorithm is added based on conventional BP neural network to prevent over-fitting of network training. In the details of selecting the node parameters of the network layer, the number of nodes in the output layer is simplified and the training parameters of the network are reduced. Finally, through the recognition experiment on the printed character samples, the recognition accuracy reaches 98.97%, which proves that this method can effectively improve the character recognition rate.
| [1] | 黄方. 人工神经网络的字符识别[J]. 电子测试, 2020(16): 76-77+95. |
| [2] | 张成, 戴俊峰, 熊闻心. 融合LeNet-5改进的扫描文档手写日期识别[J]. 计算机工程与应用, 2021, 57(9): 207-211. |
| [3] | 王逸铭, 于永洲, 孙光峤. 基于神经网络模型的扫描电镜图像字符识别方法[J]. 制造业自动化, 2020, 42(7): 18-20. |
| [4] | 柴伟佳, 王连明. 一种基于混合神经网络的车牌字符识别方法[J]. 东北师大学报(自然科学版), 2018, 50(1): 63-67. |
| [5] | Liu, M., Jiang, Q. and Hu, J. (2021) Detection of Highway Lane Lines and Drivable Regions Based on Dynamic Image Enhancement Algorithm under Unfavorable Vision. Computers and Electrical Engineering, 89, Article ID: 106911.
https://doi.org/10.1016/j.compeleceng.2020.106911 |
| [6] | Huang, Z., Wang, Z., Zhang, J., Li, Q. and Shi, Y. (2021) Image Enhancement with the Preservation of Brightness and Structures by Employing Contrast Limited Dynamic Quadri-Histogram Equalization. Optik, 226, Article ID: 165877.
https://doi.org/10.1016/j.ijleo.2020.165877 |
| [7] | Qi, Y., Yang, Z., Lian, J., Guo, Y., Sun, W., Liu, J., Wang, R. and Ma, Y. (2021) A New Heterogeneous Neural Network Model and Its Application in Image Enhancement. Neurocomputing, 440, 336-350.
https://doi.org/10.1016/j.neucom.2021.01.133 |
| [8] | 胡钰, 李甜甜, 黄梁松, 李玉霞. 基于双边伽马校正的保亮度图像增强方法[J]. 计算机应用与软件, 2019, 36(5): 204-210+241. |
| [9] | Singh, S., Sarangi, P.K., Singla, C. and Sahoo, A.K. (2021) Odia Character Recognition System: A Study on Feature Extraction and Classification Techniques. Materials Today: Proceedings, 34, 742-747.
https://doi.org/10.1016/j.matpr.2020.04.680 |
| [10] | Chen, H., Wang, Y., Zuo, M., Zhang, C., Jia, N., Liu, X. and Yang, S. (2022) A New Prediction Model of CO2 Diffusion Coefficient in Crude Oil under Reservoir Conditions Based on BP Neural Net-Work. Energy, 239, Article ID: 122286. https://doi.org/10.1016/j.energy.2021.122286 |
| [11] | Qin, H., Yang, R., Guo, C. and Wang, W. (2021) Fault Diagnosis of Electric Rudder System Using PSOFOA-BP Neural Net-Work. Measurement, 186, Article ID: 110058. https://doi.org/10.1016/j.measurement.2021.110058 |
| [12] | Wang, J., Wen, Y.Q., Gou, Y.D., Ye, Z.Y. and Chen, H. (2017) Fractional-Order Gradient Descent Learning of BP Neural Networks with Caputo Derivative. Neural Networks, 89, 19-30. https://doi.org/10.1016/j.neunet.2017.02.007 |
| [13] | Li, Z., Qin, Y., Hou, S.D., Zhang, R. and Sun, H.X. (2020) Renewable Energy System Based on IFOA-BP Neural Network Load Forecast. Energy Reports, 6, 1585-1590. https://doi.org/10.1016/j.egyr.2020.10.048 |
| [14] | 张永宏, 吴鑫. BP神经网络在图像字符识别中的改进和应用[J]. 南京信息工程大学学报(自然科学版), 2012, 4(6): 526-529. |
| [15] | 杨济萍. 基于主成分降维模型的手写数字识别研究[J]. 网络安全技术与应用, 2021(3): 31-32. |
