Wanli Chi

Electronic imaging using a logarithmic asphere

Transmission functions are derived that are valid in the nonparaxial case for a class of lenses that will image a continuum of points along an optical axis to a single image point. This lens, which we call a logarithmic asphere, is then used in a digital camera. The resolution of the camera is limited by the pixel size of the CCD; i.e., it is not diffraction limited. Digital processing is used to recover the image, and image-plane processing is used for speed. We find a tenfold increase in the depth of field over that for the diffraction-limited case.

Infrared holography using a microbolometer array

We describe a series of experiments to demonstrate holography at far-infrared wavelengths using an uncooled microbolometer array. Simple interference patterns and Fresnel zone holograms are recorded with a 10 W cw CO(2) laser illumination in a Mach-Zehnder interferometer setup. A sparse-sampling method is used to sample the hologram at a rate dependent on the bandwidth of the object wavefront rather than the carrier frequency. The samples are then used to reconstruct the complex object wavefront in the hologram plane, which is Fresnel backpropagated for image reconstruction. Uncooled microbolometer arrays are most commonly used in passive mode to image the thermal-blackbody radiation. Their technology has matured to include the wavelength range of far-infrared to submillimeter radiation. The use of microbolometers with active illumination for holography, as described in this paper, suggests their interesting future applications.

Light-emitting diode illumination design with a condensing sphere

An illumination system is described that comprises a glass or plastic sphere, an embedded light-emitting diode (LED) source, and a reflecting or scattering mirror. The rays from the LED are either refracted in the forward direction or totally internally reflected by the sphere. The mirror breaks the total internal reflection and thus sends the rays in the forward direction. The illumination pattern and efficiency of the system are analyzed. This system is extremely compact and efficient and is a good starting point for a complex illumination system design.

Extending the depth of field through unbalanced optical path difference

We describe a simple method to extend the depth of field of a conventional camera by inserting a transparent annular ring in front of the pupil of the lens. The insertion of the ring creates an unbalanced optical path difference across the lens aperture, which partitions the pupil and leads to an extended depth of field. This system is analyzed by diffraction and random process theory. Experiments are reported that are in good agreement with the theory.

Incoherently combining logarithmic aspheric lenses for extended depth of field

We describe a method for combining concentric logarithmic aspheric lenses in order to obtain an extended depth of field. Substantial improvement in extending the depth of field is obtained by carefully controlling the optical path difference among the concentric lenses so that their outputs combine incoherently. The system is analyzed through diffraction theory and the point spread function is shown to be highly invariant over a long range of object distances. After testing the image performance on a three-dimensional scene, we found that the incoherently combined logarithmic aspheres can provide a high-quality image over an axial distance corresponding to a defocus of +/- 14(lambda/4). Studies of the images of two-point objects are presented to illustrate the resolution of these lenses.

Smart camera with extended depth of field

A smart camera is studied that combines the blurring logarithmic asphere lens and maximum entropy processing to extend the depth of field 10 times over that of a conventional lens and to provide near diffraction-limited performance. For this camera, a circularly symmetric logarithmic lens with radially varying focus length provides an image with distance invariant blur; a newly developed digital deconvolution technique, the accelerated maximum entropy processing, is thereafter applied to recover the sharp image providing an extended depth of field. Three types of logarithmic aspheres are described together with the effect of aperture apodizations. We show that the central obscuration of the aperture improves the overall performance of the smart camera, especially the near distance performance. The performance of the maximum entropy processing is greatly improved by introducing a new metric parameter into the algorithm. In the presentation, I will demonstrate the novel maximum entropy algorithm that has been devised, and also demonstrate the lens manufacturing technique of magnetorheological finisher that we used.

Extended depth of field using the logarithmic asphere

Transmission functions are derived for a novel logarithmic aspheric lens that provides an extended depth-of-field when used in an integrated digital processing system. Diffraction-theory equations valid in the non-paraxial case are presented together with a comparative study of two designs. This lens can be used either singly or in conjunction with a standard photographic lens. A tenfold increase in the depth-of-field has been obtained.

Illustrative EDOF topics in Fourier optics

In this talk we present a series of illustrative topics in Fourier Optics that are proving valuable in the design of EDOF camera systems. They are at the level of final examination problems that have been made solvable by a student or professoi having studied from one of Joseph W. Goodmans books---our tribute for his 75fr year. As time permits, four illustrative topics are l) Electromagnetic waves and Fourier optics;2) The perfect lens; 3) Connection between phase delay and radially varying focal length in an asphere and 4) tailored EDOF designs.

Applications of accelerated maximum entropy algorithm

We describe an accelerated maximum entropy algorithm with wide usefulness in digital image processing. Illustrations to be described in this talk include applications to extend depth of field, imaging through turbulence, and imaging through tissue. We have found that image quality obtained is substantially better than that using inverse filter.