The image quality is the most direct method for evaluating actual optical system's quality. Imaging of standard feather
image, such as resolution plate, comet and so on is also frequently used in lens test in optical industry. The existing
simulation method for optical system can be divided into two categories. One is using ideal optical system model to
simulate image the object formed, but the actual parameters, aberration, processing and assembly of the optical system
cannot be considered. The other can calculate the special field's Spot Diagram, Point Spread Function and Module
Transfer Function by the optical design software. It is widely used for optimization (design) and image restoration. In
this paper, the actual optical system is simulated by our self-developed simulation software, the target image is entered,
the ray tracing method is used and the simulated image is received, which is formed by the object image of the practical
optical system with errors. Known parameters of the optical system are used in this article. The correctness of simulation
results is demonstrated. And the targeted photorealistic images with colors and photometric information can be gotten by
the simulation software, and the more rays be used, the more fields be cut, the simulated image much closer to the actual
image. The errors possible existing in optical system processing and assembly, such as eccentric, tilt or surface error are
considered in our simulation software, and the simulated image we get is affected by these factors. So that we can give
the reasonable tolerances, speed up the development efficiency and reduce cost.
Polarization state of only completely polarized light can be analyzed by some software, ZEMAX for example. Based on
principles of geometrical optics, novel descriptions of the light with different polarization state are provided in this
paper. Differential calculus is well used for saving the polarization state and amplitudes of sampling rays when ray
tracing. The polarization state changes are analyzed in terms of several typical circumstances, such as Brewster
incidence, total reflection. Natural light and partially polarized light are discussed as an important aspect. Further more, a
computing method including composition and decomposition of sampling rays at each surface is also set up to analyze
the energy transmission of the rays for optical systems. Adopting these analysis methods mentioned, not only the
polarization state changes of the incident rays can be obtained, but also the energy distributions can be calculated. Since
the energy distributions are obtained, the surface with the most energy loss will be found in the optical system. The
energy value and polarization state of light reaching the image surface will also be available. These analysis methods are
very helpful for designing or analyzing optical systems, such as analyzing the energy of stray light in high power optical
systems, researching the influences of optical surfaces to rays' polarization state in polarization imaging systems and so
on.
The difficulty of tracing rays through the random index medium, like atmosphere, or flowing liquid, lies on how to fast
obtain the refractive index and gradient values of each location along the ray trajectory in space. Recently, the method of
"self-adapting grid" has been developed to efficiently describe such a medium. There are several available numerical
techniques for tracing rays through inhomogeneous medium, such as Taylor method and 3rd order Runge-Kutta method,
the combinations of the self-adapting grid with different ray tracing techniques result in different accuracy and require
different computational effort. In this paper, according to the fundamentals of numerical computation, the derivation of
4th order Runge-Kutta method is presented. For the convenience of comparison, the ray trace through a regular gradient
medium (GRIN) has also been made by these methods, and the results are compared with that from other commercial
lens design codes. The results show that the 4th order Runge-Kutta method has the highest accuracy for the same step
size, but it consumes the longest time. When the ray trace step is relatively large, i. e., one fifth of the GRIN rod size, the
4th Runge-Kutta method is much more accurate than the 3rd method, however, there's few difference of the accuracy
when ray trace with a small step size. Therefore, the 3rd order Runge-Kutta method is the optimum choice for ray tracing
in self-adapting grid when comprehensively considering the accuracy and computational efforts.
A grid computing in the field of optics is presented in this paper. Firstly, the basic principles and research background of
grid computing are outlined in this paper, along with the overview of its applications and the development status quo.
The paper also discusses several typical tasks scheduling algorithms. Secondly, it focuses on describing a task scheduling
of grid computing applied in optical computation. The paper gives details about the task scheduling system, including the
task partition, granularity selection and tasks allocation, especially the structure of the system. In addition, some details
of communication on grid computing are also illustrated. In this system, the "makespan" and "load balancing" are
comprehensively considered. Finally, we build a grid model to test the task scheduling strategy, and the results are
analyzed in detail. Compared to one isolated computer, a grid comprised of one server and four processors can shorten
the "makespan" to 1/4. At the same time, the experimental results of the simulation also illustrate that the proposed
scheduling system is able to balance loads of all processors. In short, the system performs scheduling well in the grid
environment.
The image is blurred and shaken when the object through a flow with random refractive index distribution. As the
response time of the image detector is far longer than that of the random disturbance, what the detector actually detects
is an integral picture within the response time. Therefore, researches on the mean image quality within the detector's
response time are great helpful to image recovering and correction.In this paper, Monte Carlo method is used to
simulate the time-related random fluctuation of the inhomogeneous medium, which can be treated statistically.
According to the mean value and variance of the refractive index at each position within the random medium during
detector's response time, many sample refractive distributions are generated to describe the disturbance with time. And
then, ray trace through the generated medium with different refractive index distribution is made, and the spot diagram
at each sample medium is summed up as the final result. Finally, taking the flowing liquid as an example, we obtained
the spot diagram through a medium with disturbance, and made a comparison with that through a medium with the
mean refractive distribution.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.