Although I do agree with the general notion of your argument, I believe in this case it is being applied too abstract.
Yes, both can be done. Yet it is not the scarcity of intellect I'm criticising. It's the researchers time that is being spent (or wasted imho) that is the focus of my criticism.
A single RFID, QR code could've worked. Would take about 5 minutes dev time.
I can't think of having to classify the images