When a duplicate image is detected, the website generates a detailed message identifying the duplicate. The message is always accurate.
The system can easily detect when a pixel identical image is uploaded and it is handled automatically by rejecting the new upload and using the existing image instead.
If you make a minor change to an image so the images are not identical pixel by pixel, the system compares the metadata for the images, including the date and time the image was created. If there is a match, we warn the user when they upload the image and ask them to resolve how the images should be handled. The matching process will not be perfect, which is why we ask authors to specify the action to be taken.
If you have a duplicate title, we warn the user, but add the Image Id to the title to make it unique. No action is required by the author.