Micro-optical multichannel imaging systems are inspired by natural compound eyes of insects and are therefore often called artificial compound eyes. These systems contain many channels that each image a part of a wider field-of-view (FOV). In most of the demonstrated artificial compound eyes, the channels have been designed with the same imaging characteristics. However, more sophisticated imaging functionalities could be added to such systems by giving the various channels different imaging properties.
In his PhD thesis, Gebirie realized a miniaturized multichannel multiresolution imaging system in which the different channels have different imaging properties, namely a different FOV and angular resolution. These imaging systems are able to capture images of different FOV and magnifications, at the same time, over different areas of an image sensor. This could allow different image processing algorithms to be implemented to process the different images. For example a motion detection algorithm could be implemented on the wide FOV and low resolution image, whereas a face detection algorithm could be used for the narrow FOV and high resolution image.
At the start of his PhD work, Gebirie designed a basic three-channel multiresolution imaging system where each of the three channels consists of four aspherical lens surfaces. He fabricated these lenses in PMMA through ultra-precision diamond tooling and afterwards assembled them with aperture stops, baffles and a commercial CMOS sensor having an array of 1440x960 pixels each sized 10 µm. The first channel possesses the highest angular resolution (0.0096°) and narrowest FOV (2x3.5°); whereas, the third channel has the widest FOV (2x40°) and lowest angular resolution (0.078°). The second channel has intermediate FOV (2x10°) and angular resolution (0.029°). Finally, he characterized the integrated imaging system in a proof-of-concept demonstration and achieved a well comparable experimental and simulation results.
The basic three-channel multiresolution imaging system was designed at a single wavelength of 587.6 nm; therefore, its operating spectral range was limited due to the effect of chromatic aberrations. To reduce the influence of chromatic aberrations, Gebirie introduced hybrid lenses, which contain diffractive surfaces on top of refractive ones, within the previous designs of the three channels. He then fabricated the hybrid lenses through ultra-precision diamond tooling, assembled and verified in an experimental demonstration. The three channels with hybrid lenses show better image quality (both in the simulation and experiment) compared to the purely refractive three channel design.
Due to the large focal length of the first channel in both the design with the refractive and hybrid lenses, the depth of field of the aforementioned basic three-channel multiresolution imaging systems was limited. Therefore, Gebirie integrated a voltage tunable lens in the first channel to extend the depth of field of the overall system. This has resulted in a refocusing channel which has been able to focus objects at different object distances by changing the voltage applied to the tunable lens. He simulated the performances of the refocusing channel and also experimentally validated the system in a demonstrator set-up. The refocusing capability has significantly improved the depth of field of the system and ranged from 0.25 m to infinity compared to 9 m to infinity for the aforementioned basic three-channel multiresolution imaging system.
Finally, Gebirie explored depth estimation functionality to the basic three-channel multiresolution imaging system by placing a microlens array at the image plane of the first channel, which resulted in a plenoptic camera system. He further processed the image obtained, namely the integral image, to calculate the relative position where the object was located. He generated the depth map of a 3D object from two superimposed integral images which were simulated for two objects centred at two different points in a 3D space.
Gebirie presented the results of his PhD work on the design and demonstration of a basic multichannel multiresolution imaging system, the expansion of the operating spectral range of the system in the visible domain, and the refocusing and depth estimation functionalities embedded into the system thus paving the way to realize low-cost multifunctional imaging systems.
VUB - Campus Etterbeek
Building F - 9th floor