Currently I am doing it manually using Google search by image:
- Search by image using some existing small image file (with sorting results by image size)
- Select biggest returned image that have thumbnail exactly the same (or nearly exactly the same) as original image - to avoid related images or changed versions
- Download it and repeat using bigger image as seed. Stop on failure to find a bigger image.
It seems to be obvious task that may be easily automated. Is there some existing tool able to do this? Or free API allowing to make it?
I have no problem with writing some code to accomplish it but I am not willing to pay money for API/tool.
It is supposed to work on Ubuntu and/or Windows 7.