multimediamike / mobycairo Goto Github PK

It would be really useful for redump if ring unwrapping could be added, as it makes reading the ringcodes way easier.

Here's how I've done it in GIMP:

1. crop scan to ring
2. in gimp: filters>distort>polar coordinates
   circle depth: 100%
   map backwards = off
   map from top = off
   to polar = off [from polar, i.e. to rectangle]
3. reduce height to 25%

Example result:

During the cropping step if you draw the rectangle big enough to go outside the bounds of the image (which is a "valid" case when the image you're trying to crop ends right at the edge) and then you switch to Save, no cropping occurs. However there is no warning that something is wrong, in case the rectangle is drawn outside the bounds it should automatically clamp to bounds or at least provide a warning that the rectangle needs to be 100% inside the image bounds.

While the Tkinter library takes into account monitor scaling when rendering fonts, that sometimes isn't adequate. Implement an option to manually adjust the font size of any text in the panel.

I mean, if it cannot rotate and crop in batches, its absolutely redundant vs the stock windows photos app.

The aspect ratio math produces a UI that typically overflows the height of a landscape-mode display. Attached is a simple example. When this sample is opened, the "Process (C)ircle" button at the bottom of the UI frame will probably be cut off.

It would be useful to see the center of the detected circle in addition to the circle.

It would be nice to develop a workflow to process a selected list of files. This would be a small improvement to further help accelerate the overall workflow.

Add an option to allow skipping the 'cropping' portion of the workflow.

Instead of just using static red lines to indicate the discovered straight lines or circle, use a pseudo-animated color rotation by updating the redrawing the same lines several times per second with different colors. This avoids the problem of the red lines possibly clashing the red in the underlying photo.