I have an application which displays N "sub-images" as a single composite image where N can be from one to six.

The images are scaled preserving their original aspect ratio. The composite image is scaled to fit the full display height or full width. The scaling factor is determined by calculating whether the composite image will be width or height constrained.

The composite image changes scale to fit additional images appended horizontally as they are acquired (maybe just one, up to 6 total). A new set of 1-6 individual sub-images is created every few minutes and the process is repeated. The previous composite image is "cleared", and a new composite image is created and modified as a new set of sub-images are acquired.

Currently, the application uses both OpenCv and Qt. The scaling of the sub-images and creation of the composite image is done in OpenCv. Then the composite image is assigned to a single Qt label's pixmap for display.

I'm doing a simplified version of the application and I'm considering using only Qt. I haven't seen an example that does what I need to do, but using only Qt I figure I'll calculate a scale factor just like I do when using OpenCv, use a QLabel for each image, and calculate a position for each sub-image (label).

Is there a better or higher level approach that would be better?

Thanks,

Dave THomas