In this project, we generate a Chinese ID Card database containing 481 images and use neural network to train a pattern recognition model using Matlab.
We use neural network to recognize ID card information.
This process of Chinese ID Cards Recognition System mainly contains five steps.
- Load Images
- Preprocess Images
- Extract Feature
- Train the Neural Network Model
- Choose Best Model to Do Recognition The main workflow is shown in figure:
For our program, the image preprocessing contains four steps. The figure below shows the flow chart of image preprocessing.