Different reported values using same badge picture with different backgrounds

publiclab / SmART-Form

A community science effort to measure formaldehyde in the home environment

https://u.osu.edu/smartform/

GNU General Public License v3.0

9 stars 5 forks source link

Different reported values using same badge picture with different backgrounds #7

Closed gretchengehrke closed 5 years ago

gretchengehrke commented 7 years ago

I don't know all the details of this, but will update the issue as I learn them. A community member downloaded the before and after pictures, and using one background, got a reading of 75 ppb, and using a different background, got a reading of 0 ppb. I do not yet know what backgrounds she used.

shapironick commented 7 years ago

Thanks GG! do you know if they were using ISO or Android and what hardware and software?

Also I'm not sure I understand what you mean by background? to you mean lighting or what the printed badge was laying on? or what was behind the monitor? Sorry for being a touch confused.

gretchengehrke commented 7 years ago

I am confused too. She said that the background with 75 ppb was just the room, and the background with 0 ppb was a white paper. I'm wondering if she might have just been taking pictures, rather than having downloaded the images and taking pictures of those. I'll call her later today to get clarification and will respond when I know more (and what OS she has).

ebarry commented 6 years ago

Is there any more information to add to this report? Or perhaps it has been resolved and we can close this?

bihechen commented 5 years ago

So the badge is not only the colored triangle, it containes a white background. We also adjust the photo after it is taken, and the background should always be white. This should not be a problem.

shapironick commented 5 years ago

I think we are not going to get any more specifics about this one-time issue as it is a year old and am going to close this issue. We could be having issues with complex lighting conditions but we need to resolve that at the level of experimentation and field validation w/ reference methods.