About two years ago, I mentioned Point Grey cameras. These are cameras sold to the machine vision and industrial inspection market, and are much cheaper than typical microscopy cameras – most are <$1000. Point Grey puts out very nice spec sheets listing all of their cameras, and the specifications for some are pretty impressive – cameras with < 3e- read noise for ~$500. Nico Stuurman has recently written a Micro-manager driver for these cameras, and was kind enough to let me test one of these cameras. We mounted it opposite a Hamamatsu Flash4.0 (an older Flash4.0, with ~72% QE), and did a qualitative comparison by taking sequential images of the same test slide on both cameras.
The Point Grey camera we tested was a Chameleon3 CM3-U3-31S4M. This uses a Sony IMX265 sensor, which has 2048 x 1536 3.45 μm pixels, with 71% QE, <3e- read noise, and sells for ~$500. It can run at up to 55 fps. On paper, this camera should perform almost as well as the Flash 4.0. The images below are of a Texas red-phalloidin stained cell, captured with a 20x / 0.75 NA objective and a 10 ms exposure on both cameras. Click on the images to see the full size image.
A couple things are immediately apparent from these images. First, the Point Grey camera is clearly good enough to capture routine fluorescence images. Second, the smaller pixel size of the Point Grey camera actually gives better resolution images with the 20x / 0.75 objective. The resolution limit of a 0.75 NA objective at 600 nm is 0.61*λ/NA = 488 nm, so Nyquist sampling with a 20x objective would require 4.8 μm pixels; the 6.5 μm pixels of the Flash 4.0 undersample, and the 3.45 μm pixels oversample.
Second, the Point Grey camera is also somewhat noisier. In part, this is due to the fact that each pixel collects (6.5 μm / 3.45 μm)2 = 3.5-fold less light. To account for this we took a second exposure where we used a 3.5-fold longer exposure on the Point Grey camera.
With this longer exposure, and the same number of photons collected per pixel, the noise between the two cameras is much more comparable. The resolution improvement for the Point Grey camera is more striking using a 10x / 0.45 objective. Here, the 3.45 μm pixels pretty much exactly Nyquist sample, doubling the resolution achieved with the Point Grey camera compared to the Flash 4.0. Here, the exposures were chosen so that the number of photons per pixel should be constant.
Based on these simple tests, it seems clear to me that these cameras can be used at least for basic fluorescence microscopy. More testing will be needed to determine how well they will work for more demanding applications. There are already some issues that are apparent. First, it’s not clear if we had the camera optimally configured for these experiments. I think that we may have had the gain set too low, introducing quantization noise. Second, these cameras are not cooled, so dark current will be a concern for long exposures. Third, it is clear that there is insufficient hot pixel correction on these cameras. Compare these images with no light reaching the sensor.
Many of these hot pixels could probably be removed by a simple median filter following mapping of the hot pixels.
The IMX250 based sensors should perform even better – they are 5 megapixels, specified with 76% QE and 2.4 e- of read noise, and 3.45 μm pixels.
While more work is needed to make these cameras perform optimally, it’s also clear that they are already sufficient for many applications, particularly ones requiring high resolution with low magnification objectives. Furthermore, the very low cost of these cameras enables experiments requiring multiple cameras or low cost instrumentation where a conventional scientific camera would have been prohibitively expensive. I’m excited to see what people do with these.