El Gamal's Research Group at Stanford University

Multi-Aperture Image Sensor

A Multi-Aperture (MA) image sensor is designed with an array of apertures integrated onto a single chip. Each aperture contains its own local sub-array of pixels and image forming optics. By focusing the integrated optics onto an image plane formed by an objective lens in a region above the MA imager, the apertures capture overlapping views of the scene. The correlation and redundancy between apertures, along with computation, provide several new capabilities. The most notable feature of this design, which motivates the use of submicron pixels, is that a depth map of the scene may be extracted along with the image. The accuracy in the depth calculations depends on estimating the locations of features within each sub-array of pixels. The positions of features rather than the features themselves may be estimated to resolution higher than a diffraction or aberration limited lens can provide. Furthermore, very high resolution sensors may be made possible because the arrays of pixels may be disjoint. This allows flexibility in readout and correction for manufacturing. Color performance is improved as neighboring pixels all contain the same filter. This design is also useful for close-proximity imaging where the objective lens can be eliminated in order to produce a flat imaging system.

Multi-Aperture Image Sensor

Faculty

Student

Funding