The project focus on building a model to recognize the handwritten text which is fed as an image to the model. The dataset used is the most popular handwritten dataset available online which is IAM dataset for words. It consists of plenty of handwritten words from different authors. CRNN and AlexNet models have been built. CTC loss function is used for both the models. Accuracy obtained for CRNN on test samples is 73.44% and on AlexNet is 34.37% The project is being extended to build a new model for comparison. VGG16 is implemented which is giving an accuracy of 65% on test set.
-
Notifications
You must be signed in to change notification settings - Fork 0
Anusha-Manjunath-03/Handwritten-Text-Recognition
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
IAM Database: https://fki.tic.heia-fr.ch/databases/download-the-iam-handwriting-database
Topics
Resources
Stars
Watchers
Forks
Packages 0
No packages published