Hello! I’ve been experimenting with Blynk in order to gauge suitability for a new project I am working on and I’m looking for suggestions on how to make a feature work. I have a device that uses Blynk for control and status and also takes images periodically with the goal of saving them, creating time lapses, etc. I would like to upload these to some sort of cloud storage for archiving. Ideally I would not have to do much to manage this, so I was thinking users of my device could enter their google drive or dropbox credentials into their device and it would be able to push the photos there, but upon investigating this it looks like it is a bit harder to do than that since those sites require OAuth 2 login, which seems a bit heavy for a IoT device. Another option I was thinking of that would be super cool would be if Blynk could integrate with AWS S3 buckets or some other file storage service and then also provide a viewer for it. I’ve looked at the image gallery objects and those could possibly be made to work somehow but I’m not quite sure how exactly. Could I maybe have the device publish to an S3 bucket and add the Blynk AUTHTOKEN to the image name so that they can be found?
I love all the features that Blynk provides and it does a great job at making the control and data interface with my product, but it would be really complete if it could also do the media viewing portion. I’m open to all ideas on how to make this work! I appreciate and suggestions the community has. Thanks!
As @John93 says, you don’t need the auth token to the file name, unless you want to use that as a way of organising the images. But, be aware that if you use dynamic provisioning (Edgent) then the auth token will change if you re-provision, so device name may be a better variable to use.
The image gallery widget simply needs to know which publicly resolvable location to go to to fetch the images.
Obviously Blynk won’t be able to create time lapses, and is guess you’d need to use the video stream widget to display timelapse videos.
Presumably you’ll use some sort of server side image processing on your web storage location to do this, so you may need to set up a hosting service with some sort of batch processing script to assemble your time lapse videos from the still images.
Thanks for the info. Yeah I was just considering adding the auth token or some other unique ID to the image name so that they could be organized and searched.
That seems like it could work, but then the next thing I think I may need to worry about is how to handle security. This is a commercial project so I can’t just make all these urls publicly available. How would I add security to this concept?
They have to be publicly resolvable.
You may be able to add some sort of authentication, but I’m not sure that would add very much.
The URLs to the images will be of your making, so could be as complex and obscure as you wish.