For digitizing documents, traditionally, we had to use expensive hardwares like Scanners. With the advent of machine learning we can now just use our phone camera to scan and digitize all kinds of paper documents. Mobile document scanners are a must-have for students and businesses for removing busy background and generating high resolution perfectly aligned images.
Several document scanners were readily available in the market until recently when the Indian government banned most of these apps due to privacy concerns since personal data being scanned by these apps were going to private firms. Hence an open-source mobile document scanner becomes a need of the hour.
Required skills: Python, Deep Learning, Image processing, OpenCV
Bonus skills: Cloud computing, Docker, Tesseract OCR
Possible mentors: Rajwrita Nath, Rishiraj Acharya, Ritesh Prasad Singh
References and open-source building blocks:
- Models: Tensorflow, Keras, Scikit-learn, TensorflowJS
- Virtualization: Docker
Points to consider in the proposal:
Design of the system:
- What are the inputs, what are the outputs?
- What are the package requirements?
Performance:
- How long does it take? What are the time limitations?
- What resources can be used to speed it up?
- Is there associated cost?