Inspiration
As students, isn’t it common for your phone’s gallery to be filled with pictures of tutorial answers. Life would be so much more convenient if we could convert the writing in those pictures to text in a word document straightaway.
What it does
Our app takes a picture and runs it through an Optical Character Recognition (OCR) library to identify the words in the picture and output it as text on your smartphone, which can be saved as a text file. While most image to text applications convert the image to a pdf file, our app converts it to text that users can edit.
How we built it
We used Android Studio to develop our app, using OCR to convert text in images to editable text format.
Challenges we ran into
The program is not perfect and sometimes, the text conversion would not be 100% accurate. For example, there is a possibility that a ‘T’ could become a ‘Y’. We aimed to solve this problem by comparing the output text against a dictionary word by word and highlighting the possible mistakes to inform the user.
The formatting of the text would also be an issue, especially if the text in the photo are not in paragraphs and instead are in point forms (like PowerPoint presentations). However this is more of a constraint of the OCR and there was little we could do about it with the limited time.
Accomplishments that we're proud of
Completing the project from scratch in 24 hours.
What we learned
How to develop an Android app using Android Studio
What's next for SnapText
We can process the image before converting it to text. For example, we can change the perspective of the image to give a bird’s eye view, or remove background noise from the image. This can be done using OpenCV.
Log in or sign up for Devpost to join the conversation.