I want to vectorize an old map.
So I am using image trace.
I tested various settings of image trace on a small, cropped version of the image. And the results were promising.
When I applied the same settings to the entire image, the results were really poor.
This suggests to me that the applied algorithm is affected by the size of the initial image. So the same pixels will result in different vectors if the rest of the image is larger.
My options may be...to chop of the image into a bunch of smaller sections, vectorize, and recombine.
So, the questions: 1. Is my assumption correct, or perhaps I am missing something? 2. Is there a setting to override this change? 3. Does anyone have a better solution?