Author: Lucas Ekeroth


Published on IPQ website: 9th March 2015
Provide real-time ”super-resolution”, ”resolution enhancement” function in an IR-camera, in live or post-processing mode, where it is not possible to achieve that functionality for the whole image due to computational / bandwidth limitations.

Users of our IR cameras often need higher resolution in the image (both in live and for saved images).

We have for some time been working to introduce a super-resolution function for our users and to keep pace with competitors. In connection with this work, it becomes clear that a live implementation of this function becomes excessively heavy computation and is therefore impossible at present, to implement this function for live and post-processing mode in the camera. As a consequence of that the super-resolution functionf works in the current solution by post-processing of data collected in the camera but computed / processed outside the camera.

In order to meet the need for the user to view high resolution image information in the IR-camera, functionality / feature that makes that possible is required.

My idea is to reduce the amount of data that needs to be processed to reach a high resolution experience similar to the super-resolution function. By just process the image data in a sub-window of the entire image, the amount of data is reduced to get the high-resolution information. This high-resolution window can then be presented in / on regular full shot image, on a display / corresponding output.

Also sub-window post-processing of the super-resolution function collected data in the camera can be processed similarly in the camera or as in today’s super-resolution function, outside camera.

The data needed to calculate / process the high-resolution information, taken from for example a ”circular buffer” of x full or sub-window images, is stored when saving an image in a ”jpeg tag” of the saved image. This enabled our application software to handle full or sub-window resolution enhancement. You can then turn on/off this sub-window and depending on how much data that is stored you can change the size and position.
For “general jpeg viewer-applications” software the high-resolution sub-window will be seen if it was visible at the time of the saving (WSIWG).

The sub-window size can be changed and moved in the full shot. How big the window can be is depending on how heavy the computation for the high resolution window will be, to obtain optimal experience in the given camera system.

The camera or software application zoom function allows the user to zoom in and clearly see / measure on details with greater resolution requirements. The zooming makes it also possible for the user to see details with higher resolution than the display itself in the sub-window, when the display resolution is lower than the sub-window resolution achieved with the above described function similar to the super-resolution function.

The advantage is to achieve a high super-resolution function experience that is less computation / process heavy than the corresponding function in full screen mode. That enables the user to already in live camera mode and in the field to take advantage of the high resolution data without having to post process the data outside the camera, as today’s computing heavy variants enforce.

Wider field of view lenses might be possible to use for “long” distance applications due to the ability to use zoom function to zoom in for details that otherwise would have required a narrower field of view lens.