A brighter, lighter and healthier world
Home » Education Photonics » Master theses » Multi-channel, multi-resolution smart imaging system

Benson Bo-Jiun Chen

Of my time in Belgium, the thing that stands out the most is my internship with TE Connectivity (now Commscope). Also the international aspect was great.

Multi-channel, multi-resolution smart imaging system

Description

Animals have different kinds of eyes, depending on their category in taxonomy. Unlike mammals, insects, which are very small in size, do not have a single large eye but instead many tiny eyes in a small volume. These types of eyes are called compound eyes and are an inspiration for multi-channel imaging systems of low form factor. Multi-channel imaging systems have many optical channels where each channel consists of two to four lenses and captures a small portion of the total field of view. Thanks to the current micro-optics technology, such imaging systems are able to be fabricated at wafer scale; which means that the lens systems are directly processed above the image sensor wafer [1,2].

In our research group, we have been working to exploit the available channels differently so that the different channels could have different imaging properties (field of view, focal length, angular resolution). This work is part of a strategic research project funded by IWT and occurs in collaboration with IMEC, KUL and Hogeschool Gent. At B-PHOT we have succeeded to design a three channel imaging system which has three optical channels (see Fig.1.a). The optical channel with the highest resolution is able to image fine details of a certain region of interest; whereas the channel with the largest field of view is able to image the surrounding region at low resolution. The total system (Fig. 1.b) therefore has the potential to monitor a wide FOV for cognitive purposes while giving highly resolved images. The remarkable feature of such an imaging system is that it allows to implement different image processing algorithms at the different segments of the image sensor. For example, a motion detection algorithm can be applied on the image sensor segment with the largest field of view channel while on the other hand a face detection algorithm can be applied on the highest resolution channel.

The three optical channels were designed separately with CODE V optical design software and integrated together with a CAD software (SpaceClaim® 2011).The imaging system consists of two lens arrays where each array contains three lenses as shown in Fig.2. These lens arrays are fabricated using diamond tooling out of  two plastic PMMA plates [3].

       

Fig. 1. (a) Image sensor division among the three optical channels. (b) Design of a three-optical channel imaging system.

Fig. 2.Two lens arrays in a three-optical channel imaging system.

Goals

The goal of this thesis is to optically characterize the fabricated lens arrays and check if the experimental performance matches with the simulated performance. The characterization of the components can be done with the advanced measurement tools available at B-PHOT. More precisely, the roughness of the lens surfaces, the MTF, FOV and focal length of the imaging system will be measured experimentally and compared to the corresponding simulation results we have available. The results of the experimental work will also be a vital feedback to determine the optimal way of fabricating such a multi-resolution imaging system. In conclusion you can tackle in this master thesis many technological and experimental challenges in the field of optics and imaging sensors.

Link to webpage or article related to the subject :

1. J. Duparré and F. Wippermann,“Micro-optical artificial compound eyes,” Bioinsp. Biomim.  1, R1-R16 (2006).

2. E. Moens, Y. Meuret, H. Ottevaere, M. Sarkar, D. Bello, P. Merken, and H. Thienpont ,“An insect eye based image sensor with very large field of view,” in  Proc. SPIE, Vol. 7716 – Micro-Optics , 771685, 2010

3. G.Y. Belay, Y. Meuret , H. Ottevaere,  H. Thienpont, “Optical design of a multi-channel, multi-resolution imaging system ”, in Proc. SPIE, Vol. 8429-Optical Modeling and Design , 8429-10, 2012

Contact(s)

Contact Affiliation
Heidi Ottevaere Promotor
Youri Meuret Promotor
Gebirie Yizengaw Supervisor
B-PHOT cooperates with EYESTvzw for STEM-projects in Photonics
STEM

Click here for more information...

Contact

Hugo Thienpont
Prof. Dr. Ir.
Hugo Thienpont
Managing Director
Nadia Cornand

Nadia Cornand
Secretary
ncornand@b-phot.org
+32 2 791 68 52

How to reach B-Phot?

VUB - Campus Etterbeek
Pleinlaan 2
1050 Elsene
Building F - 9th floor

More info