TRA-CAPTCHA-spider

TRA(Taiwan Railways Administration) ticket booking CAPTCHA spider

package used

1. openCV
2. BeautifulSoup
3. numpy

Step

1. spider TRA ticket booking CAPATCHA
2. remove img noise
3. cut each img of number
4. turn each img
5. resize each img
6. save each img to folder

future

• build SVM model used by scikit-learn to recognize CAPTCHA
• build neural network to recognize CAPTCHA