Scripting Question - Start by Loading PNG Info

[Shoot2Skoot]

Hi,

I'm pretty new to python, but I've been messing around with it for Stable Diffusion scripts, specifically for this webui.

For a particular prompt I'm using, I've found that running the image with the same settings at a higher resolution with a denoising strength of about 0.75 really improves the end result. Furthermore, I've found if I resize less than 200px in one direction at a time, the image maintains it's original design really well. This prompt is a character generator and I want to develop a process as such:

1. Run a batch of 2k-4k images using wildcards.
2. Select the best images.
3. Batch upscale in the img2img tab as described above.

I'm looking to create a custom script that does the above. I'm fairly confident I can figure out all the img2img goodness from other scripts, but what I can't seem to find is how to get the PNG info from the file loaded via img2img batch and use that as a starting point.

Does anyone have any guidance on where I could look? Other scripts that do this, core modules that factor into this, etc?