Medical image processing
I have a medical image dataset and want to resize the object within the image while keeping the entire image size the same as it was before resizing the object. As shown in the attached figure to visualize it.
Is there any suggested method or technique?
Hello,
I think that this straightforward approach could easily meet your requirements:
1. Reduce your input image scale with a standard affine transform (e.g. bilinear).
2. Start with a black background output image.
3. Insert your scaled image centered in the output image.
Regards,
Alvaro
You can try performing a resampling step (for 50% resizing ratio; take a pixel and ignore the other) for only the non-zero pixels (pixels don't belong to the background). Then the output resampled image could be placed on the original black background of the same size. If you would like to double the size of the image you should consider an additional step of interpolation (bilinear, bicubic, or nearest).
Do you want to want to use a image processor like Photoshop or a program like MATLAB? With Photoshop it is done as follows : Select the background color same as the image background via color palette, select the image background using magic wand, select inverse (now the foreground object is selected), choose Free Transform and resize as required which resizes only the selected object. The extra space is filled by the background which has the same color as image background as selected earlier.
- Badges
- Science topic
- Similar topics
- Computer Science
- Images