I am a NOVICE in image processing!
I am working on a science project which involves capturing multiple small regions of interest (ROI’s) within a larger FOV. I then want to stitch all the images together to cover the entire FOV of the sensor without losing any pixels. The stitching must be pixel accurate. The camera I am using is genuine monochrome and has NO Baier grid.
We have to do it this way as each region of interest is illuminated by its own deicated (special) light source.
With this project we generate a large number of images with each image being 64x64 pixels in size.
The complete (stitched together) image consists of a mosaic of 6,400 images ie has a total of 26.2MP.
I would appreciate any suggestions and/or recommendations.
This sounds like appending images horizontally and/or vertically.
ImageMagick can do this, not sure how pixel-perfect this will be though. I did do this, but the level of accuracy needed wasn’t on a single pixel level. You need to try/experiment with it. Anyway…
ImagemaGick → convert and the append option
Example:
vertical
convert -append 1.jpg 2.jpg out.jpg
horizontal
convert +append 3.jpg 4.jpg out.jpg
1 Like
Will try that.
Thanks, bedankt
Of course, stitching without overlap means that your indivudual images must be perfectly aligned on the overall grid. And that implies that your camera(s) must be perfectly aligned…
The whole process might be a lot simpler if you can record larger areas to allow aligning, and then mask to retain only the region of interest for stitching.