Document and Handwriting Digitizer Used AI

Fixed issue 500 if no text in image

Shipped project and fixed all issues. Notably - Made upload page more intuituve, you have to upload a file to continue

I significantly improved the accuracy of my website. Firstly, the website is MUCH BETTER at converting handwriting since switching models, and runs of the clou, allowing it to be locally, much smaller (and much easier for me to deploy). It also has SIGNIFICANTLY ENHANCED TEXT DETECTION, HANDWRITING DETECTION... Really seems like a finsihed product now.

Finished making the site responsive with media queries for all (or mostly all) devices.

Added a new minty blue background and got the pdf output page finalized.

Website is finished! Now Im just scaling it for mobile and adding finishing touches (well - buttons)!

Made the website display a PDF after a doc has been scanned.

I added a landing page with some usefull links and moving text.

Just added a landing page. I want to make it really cool with animations and stuff. Also added animation to the upload page.

The website at this point is basically finished.
The bugs I stomped on included:
Got dropzone working fully working.
Included a max file upload limit of 1 file
Restricted the file types to only JPEGs, and PNGs, and JPGs
Checked to see if scales properly on iphones.

The last step is adding a download PDF button for the converted file, and making the converted file show up a bit nicer.

Added uniformity to all text fonts. Next steps - Integrate hamburger menu with options like suggestions, collarborate on github...
also need to make max file uploads 1.
and still working on accepted file types

I fixed the issue of files not being able to be uploaded again if the page was reloaded.

Completely overhauled the aesthetic of a website to match the summer of making theme!. Also identified the following bugs:
1. clicking convert button when no files uploaded causes website to crash.
2. Any file format can be uploaded. Not just jpeg or png.
3. More than 1 file can be uploaded. Max files allowed should be 1.
Almost done. After these bugs are resolved, the website should be good for shipping.

Added dropzone for dragging and dropping files. I troubleshooted the issue of dropzone not uploading files to be proccesed by the OCR. Tried making the website look better, however its still a work in progress. My next steps will be improving the aesthetic of a website.

Working on making the website more visually appealing.

Managed to display the converted text onto the main display page. Still working on making the website more visually appealing though and making the output text more presentable and downloadable.

Finally! I was able to!
1. Got the Flask website working.
2. Delete files after processing so they dont hog storage space.
3. Open the converted text in a new page.

Next steps is to beautify the site, make the converted text appear as a file, that is displayed on the website and can be downloaded.

Switched to using flask instead of bottle. Improving text recognition

Created a lightweight Bottle webpage that is currently only running of my Mac. Significantly improved file selector with HTML, instead of relying on TKinter. Now working on returning the selected file to python program to analyze.

I added a TKinter GUI to select an image file to extract text from rather than having to enter a file directory to access the image. Makes it much faster and more efficient for the end user. Next step is getting a simple website running.

Basic setup. Convert a local image file on my computer to text displayed in the output window.