The facetVISION array camera architecture allows to reduce the z-height of mobile camera modules to 50% of that of a conventional camera module independent of pixel size and resolution while at the same time providing depth information. It employs folding of the optical path via a mirror bank and segmenting the field of view into different imaging channels using individual inclinations of the mirror facets as well as stitching of the final image by depth-based image rendering. A typical arrangement incorporates four channels in one line with D-cut objectives and about 2:1 aspect ratio image sensors. One pair of cameras captures the upper part of the field of view providing depth there and the second pair accordingly for the lower part of the total field of view. As an option, the mirror bank is rotated, so that the same camera module can be used to capture world-side images as well as selfies sequentially. In demonstrator systems realized by us, we used different combinations of voice coil motors with piezo bending actuators for auto focus and optical image stabilization. Tunable lenses were employed for channel-individual focusing to account for back focal length variations within the array as well as possible changes of the same, especially with respect to temperature. In this paper, we report of the in-detail analysis of the thermal effects within our array camera architecture including tunable lenses. MTF-through-focus curves are captured and evaluated with the camera being exposed to different temperatures and voltage settings for the adaptive lenses.
|