Yesterday I went back and looked at some of the simulation results, and grew suspicious that there were problems with the way the simulated camera was sampling the simulated target. As I dug deeper, I started to question some of my original assumptions, and started experimenting with changes. What I ended up with is close… [Read More]

In a comment to a recent post, Jack Hogan posted this link. It’s well worth reading. In it you’ll find Frans van den Bergh’s explanation of how a 4-way Lithium Niobate beam-splitting anti-aliasing (AA) filter works, and what convolution kernel simulates one that has 0.375 pixel shifts. I added that kind of AA filter to the… [Read More]

In the last post I showed MTF results from a simulation of a Bayer-CFA camera with a diffraction-limited lens. The camera had a 4.77 micrometer (um) pixel pitch. I had previously done an MTF analysis of the Zeiss Otus 55mm f/1.4 lens on a Sony alpha 7R, a camera with that pixel pitch. I thought… [Read More]

I’ve been working on camera and lens simulation for a couple of weeks, and I decided that I needed a technique for calculating the spatial frequency responses (SFRs) and modulation transfer functions (MTFs) of synthetic images that was a) completely open, so I could understand just what it was doing, and b) embeddable in my… [Read More]