Free GlideApp Museum translation app developed with OCR

After the Art Gallery App Free GlideApp for developing Art Gallery translation mobile app I wanted to extend it to a Museum app.

As it does not take too much time to mock up an app I thought a specific app targeted at a specific Museum would capture more attention than if you were to do a generic one. They can see the potential in one of their exhibits.

link to the app is HERE

As it does not take too much time to mock up an app I thought a specific app targeted at a specific Museum would capture more attention than if you were to do a generic one. They can see the potential in one of their exhibits.

I went along to the museum and took a few quick photos with my mobile phone, not the best pictures as the exhibits had low lighting, glass cases (that reflect light) and the museum was quite busy so I did not want to obstruct the exhibition for too long so fast photos.

OCR (Optical Character Recognition)

The exhibition that I chose has been there for a long time and there was not any supporting pamphlets or other information, so I had to rely on the photographic images for text.

After downloading the photos to my laptop I cropped the text images so that only one column was showing

Original photo taken by mobile camera, note the text on the column to the right. The OCR (Optical Character Recognition) programme will try to read that.
I used Win 10 Photo programme to crop and rotate image (also adjust lightness of image) to make it easier for OCR programme to work.
I use Google Keep (firefox extension) to upload image and “Grab image text” and show underneath. I can then edit/correct if required and copy the text to a spreadsheet.

After adjusting the text image with Win 10 Photo programme I used Google Keep’s OCR tool to get the text. The only real issue I have is with accent marks like macron’s , some of which convert to umlats. I will leave them for present, but would normally proofread to correct. This is just a demo at this point.

Framing the app

The display’s have story labels underneath. This has to be re-framed for the app. I am envisioning this to be a supportive tool for the main exhibition. Giving more insight into an object by translating the information. So the app would be used when viewing the exhibits.

There is quite a bit of text at the beginning of the exhibition, so visually a bit dull

I’ve tried to break the text up with images.

For accessing the app I have used a map with some translations pointing to a QR code on a map. Scanning the QR code will give a link to the map. An alternative method is to have a link on the main museum site which would take you to the specific app.

GlideApp new choice feature

While developing the app a new element in the GlideApp arsenal turned up, choice, there is a video on it here. It is an interesting feature but does not allow for a residual choice that can be created by multiple people and stored.

I am thinking of how you could use the app to give feedback and the “Comment” component is the only one I know of. Also, to date, I have not found a way to translate the text for this into other languages, and I do not want to put a large string of text above it as it might overcrowd a small screen.

Also comment at the bottom means you ave to get to the end to see that you can leave a comment, it is a bit hidden. I’m wondering if there is a way to make it more accessible.

End thoughts

Similar process to the City Gallery app but the OCR and using Google Translate on the whole document sped up the process.

I’m still not sure on sequence, whether to mock up the app with just English translation, then go on and translate the other languages, or translate first and then make the app. In this case I did translation first, so lots of cutting/pasting in the spreadsheet. The next one I may try doing English only first, it may be faster.

I’m becoming more familiar with the setup process. The images took a bit of time to crop/manage on this app as the displays held quite a number of items. I’m not sure I did it in the most effective manner but that is all part of the learning. I would take individual shots of more of the pieces rather than just the display overall.

This app you had to scroll down quite a lot, I’m not sure whether it would have been better to split it into other pages. That may be worth investigating too, as it will reduce the chunk size for viewers to absorb, but may end up being more confusing with way-finding through the different elements.

At the end, I’m reasonably happy with the results.

Add a Comment