Understanding Mega Pixel Resolution

A pixel is the smallest light sensitive element on an image’s sensor. They come in many sizes and sensitivities. Their function is to convert incident photons into electrons that can be measured by subsequent circuits.

Consumer marketing has hyped “mega pixels” as the hook on which they hang their hat. In scientific imaging, the number of pixels is only part of the story, in fact, not always the most important part. Here are some of the important specifications of pixels and their relationship to real world scientific imaging.

Quantity: How Many Mega Pixels Are Enough?

It depends on the optics. The non-intuitive reality is that higher magnification objectives require fewer pixels. The physics of light produces blur circles at the sensor focal plane that are dependant on the design, magnification and NA of the objective. The following table provides the results for a monochrome sensor, a typical 16 mm field of view and pixel pitch equal to 1/2x the blur circle (2x sampling frequency).

Type Magnification NA Horizontal Vertical Megapixels
Achromat 4x 0.10 1,471 1,103 1.62
Apochromat 4x 0.20 2,943 2,207 6.49
Achromat 10x 0.25 1,471 1,103 1.62
Apochromat 10x 0.45 2,648 1,986 5.26
Achromat 20x 0.40 1,177 883 1.04
Apochromat 20x 0.75 2,207 1,655 3.65
Achromat 40x 0.65 956 717 0.69
Apochromat 40x 1.00 1,471 1,103 1.62
Achromat 100x 1.20 706 530 0.37
Apochromat 100x 1.40 824 618 0.51

Next, we must take into account how microscopists use different magnifications. Typically, lower magnifications are used for surveying the specimen and higher magnifications are used for resolving the fine details. This predicates using a camera that satisfies the demands of the medium and higher magnification objectives rather than the lower magnifications.

What are some of the drawbacks to exceeding these needs?

Speed
Each pixel takes time to read out, so increased resolution slows the frame rate. This will affect your high speed sequence capability.
Computer Capacity
Large image files quickly use up the random access memory, forcing the computer to use memory swapping. This along with increased image processing time can slow down your work-flow.
Sensitivity
Sensitivity of each pixel is directly proportional to its area and pixel area is inversely proportional to the total number of pixels on the chip. Therefore, the greater the number of pixels, the lower the sensitivity of each pixel.
Pixel Capacity
The maximum number of electrons a pixel can hold is generally proportional to its area. The smaller the pixel, the fewer electrons it can hold. This reduces the dynamic range, the ability to detect dim details and bright details in one image.
The Cost of Cost Reductions
The race to reduce costs has added an additional factor to the image quality loss. Many cameras are being released with consumer grade sensors.
Pixel Size
Consumer sensors are typically smaller to reduce costs.
Noise Performance
Consumer sensor and circuit noise levels are ~5-10 times greater.
Sensitivity
Consumer grade sensors typically have half the QE2 as scientific sensors.
Quantum Efficiency
the probability that a certain wavelength photon will produce a captured electron in the pixel)
True Value
Understanding your individual needs and buying a balanced system is more important than purchasing “the biggest bang (mega pixel) for your buck.”