livezingy / merged-characters-segmentation
Algorithm implementation of the "The Robustness of “Connecting Characters Together” CAPTCHAs". The test data set is the TAOBAO CAPTCHA that composed of blue character and white background.
☆27Updated last year
Alternatives and similar repositories for merged-characters-segmentation:
Users that are interested in merged-characters-segmentation are comparing it to the libraries listed below
- Build an Optical Character Recognition service using deep learning method☆54Updated 8 years ago
- hack captcha with CNN☆37Updated 8 years ago
- Machine Learning 2016 final project☆45Updated 8 years ago
- A tools can generate samples for OCR trainning. 用于OCR的字符样本生成工具☆65Updated 7 years ago
- 验证码识别,该模型是基于xlvector模型上进行加工,验证码内容包含了大小字母以及数字,采用lstm+warp-ctc+cnn达到不分割字符而识别验证码内容～☆113Updated 8 years ago
- 代码讲解部分请前往blog：http://lan2720.github.io/☆34Updated 8 years ago
- chinese character generator and training using keras.☆40Updated 8 years ago
- A Simple Chinese OCR from tipdm contest☆64Updated 8 years ago
- recaptcha with lstm and mxnet☆29Updated 8 years ago
- ☆80Updated 8 years ago
- This software implements the Convolutional Recurrent Neural Network (CRNN) in tensorflow☆53Updated 7 years ago
- crnn in mxnet.can train with chinese characters☆48Updated 6 years ago
- ☆39Updated 7 years ago
- End-to-end variable length Captcha recognition using CNN+RNN+Attention/CTC (pytorch implementation). 端到端的不定长验证码识别☆116Updated 7 years ago
- 针对不定长的验证码使用ctc_loss+lstm进行训练，然后在测试集合上面进行测试☆18Updated 7 years ago
- Geetest Captcha Crack 为了不被怪兽吃掉而奋斗！☆22Updated 7 years ago
- Caffe for CC4.0-Windows，简单的Caffe C++接口，方便简单☆161Updated 5 years ago
- use STN+CNN+BLSTM+CTC to do OCR☆93Updated 6 years ago
- An implementation of the Fully Convolutional Recurrent Neural Network (FCRN) Framework in Keras as described by Ankush Gupta et. al in th…☆12Updated 8 years ago
- Learning Deep Learning☆226Updated 7 years ago
- ☆19Updated 8 years ago
- 基于caffe的LSTM OCR案例，能够利用该案例完成序列的识别，包括验证码、车牌、身份证号码、地址等长序列动长的内容识别☆212Updated 7 years ago
- This repo contains code written by MXNet for ocr tasks, which uses an cnn-lstm-ctc architecture to do text recognition.☆50Updated 7 years ago
- 基于传统多标签的定长验证码识别和基于GRU+CTC的不定长验证码识别☆16Updated 5 years ago
- an n2n ocr for qq captcha, 端到端的腾讯验证码识别☆86Updated 7 years ago
- Using adaboost to detect carface☆22Updated 9 years ago
- this is an implementation of alexnet with tensorflow. Run the train.py for training, and test.py for testing. You only need to config the…☆25Updated 8 years ago
- ☆15Updated 6 years ago
- ☆31Updated 7 years ago
- Text Detection System with MSER , SWT and Text Verification☆109Updated 9 years ago