<abstract xmlns="http://www.w3.org/1999/xhtml"><p>In this paper, the difficult problem of character recognition in natural scenes caused by many factors such as variability of light in the natural scene, background clutter and inaccurate viewing angle, and inconsistent resolution. Based on the deep learning framework PyTorch, a convolutional neural network is implemented. Based on the classic LeNet-5 network, the network optimizes the input layer to accept three-channel images, changes the pooling method to maximum pooling to simplify parameters, and the activation function is replaced by Rectified Linear Unit with faster convergence. The cross-entropy loss is used instead of the minimum mean square error to mitigate the slow learning. Furthermore, we also enroll the gradient descent optimization algorithm RMSprop and L2 regularization to improve the accuracy, speed up the convergence and suppress the over-fitting. The experiment results show that our model achieved an accuracy of 92.32% after training for 7h24min on the street view house number(SVHN) dataset, effectively improving the performance of LeNet-5.</p></abstract>

In this paper, the difficult problem of character recognition in natural scenes caused by many factors such as variability of light in the natural scene, background clutter and inaccurate viewing angle, and inconsistent resolution. Based on the deep learning framework PyTorch, a convolutional neural network is implemented. Based on the classic LeNet-5 network, the network optimizes the input layer to accept three-channel images, changes the pooling method to maximum pooling to simplify parameters, and the activation function is replaced by Rectified Linear Unit with faster convergence. The cross-entropy loss is used instead of the minimum mean square error to mitigate the slow learning. Furthermore, we also enroll the gradient descent optimization algorithm RMSprop and L2 regularization to improve the accuracy, speed up the convergence and suppress the over-fitting. The experiment results show that our model achieved an accuracy of 92.32% after training for 7h24min on the street view house number(SVHN) dataset, effectively improving the performance of LeNet-5.

Street View House Number Identification Based on Deep Learning

International Journal of Advanced Network, Monitoring and Controls

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

{"article-title":"Street View House Number Identification Based on Deep Learning"}

In this paper, the difficult problem of character recognition in natural scenes caused by many factors such as variability of light in the natural scene,...

Optimizer	parameter
SGD	lr=0.001,
Adam	lr=0.001,
Adamax	lr=0.002,
RMSprop	lr=0.001,

Subset category	Number of samples
Training set	73257
Extra set	531131
Test set	26032

optimizer	Top Accuracy/%
SGD	87.350184
Adam	89.090000
Adamax	88.955900
RMSprop	88.676000

Train sample number	test sample number	Best test accuracy	time
73257	26032	90.01229	1h24min
604388	26032	92.32483	6h17min

weight_decay	Train Acc%(e)	Test Acc%(e)
0.01	89.01538(85)	87.14659(85)
0.005	91.59397(57)	88.95590(57)
0.0025	94.51470(88)	90.01229(88)
0.001	97.21119(90)	89.70498(24)

Street View House Number Identification Based on Deep Learning

Published Online: Nov 11, 2019

Page range: 47 - 52

DOI: https://doi.org/10.21307/ijanmc-2019-058

Keywords

© 2019 Haoqi Yang et al., published by Sciendo

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11.

Figure 12.

Figure 13.

Figure 14.

OPTIMIZER PARAMETER SETTING

AUGMENTATION RESULT

OPTIMIZERS TRAINING RESULTS

RESULT AFTER DATA AUGMENTATION

RESULT OF DIFFERENT WEIGHT DECAY