ImageTextRecognition

Problem Statement

Given an input Image we need to predict the Text in the Image with a reasonable accuracy >80% (Exact match with the actual Text Labels) and should have a good letter match accuracy.

Performance Metrics

• Accuracy
• Letter Accuracy
• CTC Loss

Model Architecture

The Model used is Convolutional Recurrent Neural Network(CRNN) which is end-to-end trainable and its architecture consists of three parts:

1. convolutional layers, which extract a feature sequence from the input image;
2. recurrentlayers, which predict a label distribution for each frame;
3. transcription layer, which translates the per-frame predictions into the final label sequence

Please check the iPython Notebook for detailed analysis and implementation of this project.