Gregory E. Johnson

Wavefront coding: a modern method of achieving high-performance and/or low-cost imaging systems

This paper gives a brief introduction into the background, application, and design of Wavefront Coding imaging systems. Wavefront Coding is a general technique of using generalized aspheric optics and digital signal processing to greatly increase the performance and/or reduce the cost of imaging systems. The type of aspheric optics employed results in optical imaging characteristics that are very insensitive to misfocus related aberrations. A sharp and clear image is not directly produced from the optics, however, digital signal processing applied to the sampled image produces a sharp and clear final image that is also insensitive to misfocus related aberrations. This paper gives an overview of Wavefront Coding and example images related to the two applications of machine vision/label reading and biometric imaging. Design techniques of Wavefront Coding are unique from that of traditional imaging system design since both the optics and digital processing characteristics of the system are jointly optimized for optimum system performance.

Extending the imaging volume for biometric iris recognition

The use of the human iris as a biometric has recently attracted significant interest in the area of security applications. The need to capture an iris without active user cooperation places demands on the optical system. Unlike a traditional optical design, in which a large imaging volume is traded off for diminished imaging resolution and capacity for collecting light, Wavefront Coded imaging is a computational imaging technology capable of expanding the imaging volume while maintaining an accurate and robust iris identification capability. We apply Wavefront Coded imaging to extend the imaging volume of the iris recognition application.

Passive ranging through wave-front coding: information and application

Passive-ranging systems based on wave-front coding are introduced. These single-aperture hybrid optical-digital systems are analyzed by use of linear models and the Fisher information matrix. Two schemes for passive ranging by use of a single aperture and a single image are investigated: (i) estimating the range to an object and (ii) detecting objects over a set of ranges. Theoretical limitations on estimator-error variances are given by use of the Cramer-Rao bounds. Evaluations show that range estimates with less than 0.1% error can be obtained from a single wave-front coded image. An experimental system was also built, and example results are given.

Computational imaging design tools and methods

Systems which transform optical wavefronts into digital information for imagery or machine interpretation lack suitable design tools and methods. In Wavefront Coded imaging systems in particular, the signal processing and the optics must be jointly considered to achieve an optimal solution. Computational imaging systems have recently been designed at CDM Optics based on human interpretation of images and guided by machine (algorithmic) interpretations. CDM is generating an integrated design package called WFCDesign to provide for truly joint optimization of computational imaging systems where both physical and algorithmic goals can be jointly realized with a high degree of efficiency and accuracy. WFCDesign interfaces to a multitude of commercial analysis, design, signal processing, and simulation packages, enabling joint optimization using industry-standard tools. Methods for approaching the optimization problem including merit functions and optimizer issues are discussed. An example of a computational design with Wavefront Coding based on a digital algorithms performance (as opposed to strictly optical metrics such as spot size or aberration curves) is provided. An outline and discussion of the WFCDesign package highlights the capabilities of our flexible approach and modular architecture and provides insight into the future of computational imaging design tools.

Larger depth of focus for increased yield

Large depths of focus can be obtained by modifying the pupil function of a projector lens. An image recorder with a threshhold greatly increases the number of phase functions that can be used. Simulations are given that show that it is feasible to increase the depth of focus by a factor of two or more by making modification of the phase in the aperture stop of the projector lens. Examples show new techniques that are used to design pupil-plane phase functions to produce images for contact holes with very little change over at least twice the normal focal depth. The penalty is that the side lobes are higher than for a system without modification in the pupil plane. A larger focal depth gives considerable potential cost savings. For example, the down time for focus correction of a lithographic system and re-work of wafers can be reduced if the focal depth of the projector is increased. Using the smallest numbers that were quoted, the down time for refocus (2% on 20% of the jobs) costs

Passive ranging for acquisition of range images: applications to longitudinal vehicle control and warning systems

4M/yr./machine. The 30% of re-work that is caused by focus problems costs

System And Method For Optimizing Optical And Digital System Designs

15M/yr./machine. Expensive flatter wafers need not be used. The cost of flatter wafers, when needed (10% of the time) with current machines, is more than

Optical guidance systems and methods using mutually distinct signal-modifying sensors

17M/yr./machine. Using the lowest numbers quoted, the totals for down time, re-work, and flatter wafers alone is

Coded localization systems, methods and apparatus

36M/yr./machine. In addition, masks can be cheaper because complexity for extending the depth of focus is not needed. We show the phase modifications that were used, how the phase modifications were found, and highlight the differences between the design techniques that we used and the previous work on extending the depth of focus.

ANGULAR LOCALIZATION SYSTEM AND METHOD

A single-lens, single-aperture passive range-imaging system has been developed based on wavefront coding of incoherent images using a standard CCD camera and lens, and a phase mask. Two-dimensional range images or point-measurements of range may be derived using fast spectral estimators, enabling efficient, real-time, range determination. Theoretical limits available for the errors of the estimator provide powerful design and evaluation tools. Such a sensor could prove useful in longitudinal control systems (i.e. adaptive cruise control) by providing wide-field, real-time range data. Other applications include object detection for collision avoidance and backup- or blind-spot warning systems.