Skip to content

khushi-411-projects/air-handwriting-detection-and-recognition

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Air Handwriting Detection And Recognition

This project is divided into two parts:

  • Detection
  • Recognition

Execution:

Execute the air-handwriting-detection-and-recognnition.ipynb with color_selector.py in the same folder if you are running code on jupyter notebook or VS code or any local host machine with replacement of path of dataset and last_frame.png to where you have saved your dataset and last_frame.png.

If execution is done using Google Colab; Execute the detection.py with color_selector.py in the same folder of code using VS Code or prefered python editor and OCR.ipynb in google colab by uploading last_frame.png into your google drive and replacing the path of both dataset and last_frame.png.

Initially Training is set to True, Changing it to false will use the trained model but initially you have to train the model.

Detecting the Alphabets:

Screenshot (95)

The Output obtained:

mediahandler

Recognising the Alphabets:

Screenshot (97)

Dataset Description:

Link to Dataset

A-Z Dataset

Ploting Dataset

This image shows a histogram plot of complete dataset, representing number of entries of each label.

Screenshot (101)

Ploting Images:

  • Training Data
  • Testing Data (Shows the predicted label)

Screenshot (102)

The output obtained:

Screenshot (96)

Confusion Matrix of testing Data:

image

Confusion Matrix of training Data:

image

Performance Comparision Between RMSProp and ADAM optimizer:

test1 and train1 are the output of the ADAM while test2 and train2 are the output of RMSProp

image

Team Of:

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published