Skip to content

Machine Learning Project using Convolutional and Recurrent Neural Network

Notifications You must be signed in to change notification settings

jadhav-kunal/Captcha-Deciphering-Using-CRNN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 

Repository files navigation

Captcha-Deciphering-Using-CNN+RNN

Python Tensorflow Keras Colab

In this application, variations of complex text-based CAPTCHA images will be provided with number of letters, dimensions of images and distortions.

A framework is proposed to decipher the complex text-based CAPTCHAs.

The proposed framework is based principally on conventional steps of decoding CAPTCHAs (i.e. preprocessing, segmentation and recognition).

This project demonstrates a simple OCR model built with the Functional API. Apart from combining CNN and RNN, it also illustrates how you can instantiate a new layer and use it as an "Endpoint layer" for implementing CTC loss.

Convolutional Recurrent Neural Network

The CRNN approach identifies words using three steps:

1.A standard convolutional neural network (CNN) — the first layer breaks the image into features and is divided into “feature columns”.

2.These columns are fed into a deep-bidirectional long short term memory (LSTM) cell, which provides a sequence, identifying the relationship between the characters.

3.The output of the LSTM cell is fed into a transcription layer, which takes the character sequence, including redundant characters, and uses a probabilistic approach to clean the output.

Dataset

The images are 5 letter words that can contain numbers. The images have had noise applied to them (blur and a line). They are 200 x 50 PNGs. Dataset contains 1070 files.

dataset

About

Machine Learning Project using Convolutional and Recurrent Neural Network

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published