README.md



CS4804 Mini-Project OCR

Due days
04/24 Presentation
Coding should end before Presentation
05/01 Final Report & Submission

Presentation
A short class presentation (6~8 minutes)
Q & A (2 minutes)
Presentation:

Title
Problem statement and analysis
Use-Case scenarios
AI algorithm and model
Results and demonstration
Lesson learned
Q & A


Final Report
A maximum of 8 pages summarizes all aspects of the project:

Problem statement and analysis
Use-Case scenarios
Literature review
AI algorithm and model
Results and demonstration
Code and documentation
Lessons learned
Future work (if any)


Final Submission
Compile all files into a zip file,
including:

Presentation slides
Code (file or link)
Final report

Ensure that all links in your report and slides are accessible

Code Specifications


Language: Python


Third-party library: Install all with pip install -r requirements.txt

tensorflow pip install tensorflow

scikit-learn pip install scikit-learn

openCV pip install opencv-python


Trainning dataset:


English Handwritten Characters   Size: 13.7 MB / 25.2 MB(after unzip)

Handwritten Digits and English Characters Size: 85 MB / 545.57 MB(after unzip)


Usage
python train-ocr-model.py -d eng_dataset/dataset_mnist.csv -s -m new-model.keras -p result.png
python run-ocr-model.py -i simple_ocr_example/test_cases/hand_writing/1-image.png -m new-model.keras -s 28