This white paper is provided to DV Info Net courtesy of Sofradir EC, Inc., formerly Electrophysics, a leading developer of advanced high performance infrared cameras, portable thermography systems and advanced night vision imaging solutions.
It’s been remarkable to witness the evolution of the SLR camera, first as a sophisticated film camera, and now as a maturing digital camera with many very impressive features. The evolution continues as new features are being introduced and improved specifications launched at a steady rate! For the surveillance photographer, some of the newest features can be quite useful for portable photography and surveillance, both at night and during the day. This article describes some of the more important features of the digital SLR, or dSLR, camera and how these features improve the state-of-the-art surveillance photography.
DSLR cameras are highly valued by surveillance photographers because of their resolution, sensitivity, versatility in the field, availability of a wide variety of objective lenses and because they allow an accurate preview of framing close to the moment of exposure. Many photographers also prefer dSLRs, which when compared to compact point-and-shoot cameras, incorporate larger sensors now available with the same size as traditional film formats. These large sensors allow for similar field-of-view values to film formats, as well as their comparable sensitivity.
Left: Nikon’s D4 Camera utilizes a full-frame 4928×3280 pixel image sensor (16MP) with un-enhanced ISO up to 12800. Right: Canon’s EOS-5D Mark III utilizes a 5760×3840 pixel image sensor (22MP) in an APS-F format with un-enhanced ISO up to 25600.
Over 80% of the dSLR cameras sold today are models from Nikon or Canon. As a result of their predominance, an abundance of lenses and accessories are available for these two camera bodies, resulting in an excellent selection and remarkable pricing. These accessories include: objective lenses that are fast, lightweight, optically stabilized, autofocus, as well as night vision modules and other camera accessories that increase overall camera performance. Note that while there are other camera manufacturers that provide excellent products worthy of consideration for certain applications, only Nikon and Canon cameras are mentioned in this article.
Pixels: How Many? How Big?
Impact of Physical Array Size
Impact of Physical Pixel Size
Aperture and Exposure Time
Enhanced ISO – Higher Gain with Drawbacks
Night Vision Modules
Photography at a Distance
Pixels: How Many? How Big?
The most noticeable change in the evolution of digital SLR cameras has been the steady increase in the number of pixels that makeup the image sensor. Commercial cameras have seen the number of pixels in the camera’s sensor grow twenty times in as many years, from 1MP (1 million pixels) in 1990 to over 35MP in 2012 (see Figure 1). It appears that the array size is beginning to reach level off.
Figure 1: Sensor Arrays in dSLR cameras have grown steadily over the past 20 years.
Are cameras with larger sensor arrays better? Not necessarily. It depends on the application. There’s more to understanding the impact of larger array sizes than merely the number of pixels. There are two other important considerations that are considered here: the physical sensor array size and the physical pixel size.
Impact of Physical Array Size
The first dSLR cameras used image sensors that were significantly smaller than traditional 35mm film formats (36mmx24mm). As shown in Figure 2, today dSLR image sensors are available in several distinct sizes. The smallest image sensors are about 2/3 the size of the 35mm format. They are available from both Canon (known as APS-C format) and Nikon (known as the DX format). Canon also provides cameras with a slightly larger size sensor, APS-H which is approximately 28x19mm. The largest sensors (known as “full-frame” sensors) are now available in cameras from both Canon and Nikon and have the same size sensor as the original film SLR cameras, 36mmx24mm.
Figure 2: Today’s dSLR cameras have certain physical array size formats ranging from the full-frame sensor to a sensor having about 2/3 that size.
Below is a chart that shows the physical array size for current Nikon and Canon cameras:
For cameras having an image sensor that is smaller than full-frame, a digital crop factor has been defined (also known as the focal length multiplier or magnification factor) which can be calculated by taking the ratio of the diagonal dimension of 35mm film (43.3mm) to the diagonal dimension of the camera’s image sensor size. As should be obvious by the term focal length multiplier, multiplying the lens focal length by the crop factor gives the effective focal length of a lens that would yield the same field-of-view if used on a full-frame camera.
For the surveillance photographer, the concept of having a crop factor >1 is normally regarded as a benefit. For long distance viewing, a narrower field of view is quite desirable and the crop factor provides photographers a “boost” in long-focal-length enabling them to fill the frame more easily when the subject is far away. For example, as shown in Figure 3, the focal length multiplier for a Nikon DX-format camera is 1.5. Using a 200mm lens on a Nikon DX-format camera will deliver the same field-of-view as a 300mm lens on a full-frame camera.
However, while there are benefits for long distance viewing, the narrowing of the FOV is sometimes a disadvantage to photographers when a wide FOV is desired. For example, a 24mm lens on a Nikon DX format camera will result in the same field-of-view as a 36mm lens, possibly too narrow for the application necessitating the use of expensive ultra-wide lenses to deliver the desired FOV (16mm in this example).
Obviously, cameras having “full-frame” image sensors that are the same size as the 35mm film size, there is no crop factor (crop factor is 1).
Impact of Physical Pixel Size
In order to better understand the impact of the pixel size, compare the performance of two cameras that have the same physical array size but a different number of pixels (and consequently different pixel sizes). For this example, let’s compare the Nikon D3X (fullframe image sensor with 24MP) and the Nikon D3S (full-frame image sensor with 12MP). For the D3X, the physical pixel dimension is about 6μm whereas the D3S has pixels that are about 8.5μm, about 40% larger. In this example, if the military photographer is performing surveillance at a distance, identification of objects will be related to the number of pixels on target. Hence, if the same lens is used on the two cameras being compared, the standoff distance for the higher resolution model will be further than for the lower resolution camera. Explained differently, in order for the lower resolution camera to deliver the same resolving capability as the higher resolution camera with a 300mm lens, the lower resolution camera would either need a 420mm lens or the photographer would need to move closer to the target by about 25% of the distance.
Figure 4: Today’s dSLR cameras use sensor arrays with widely varying array sizes and pixel sizes. Larger sensor array size usually means better resolution; larger pixels usually mean better sensitivity.
As far as resolving capability goes, smaller pixels are better and the advantage goes to the high resolution camera which can use lenses that are lightweight, easy to hold and essentially “faster.”
Unfortunately, more smaller pixels aren’t always the best thing for a surveillance photographer! All things being equal, as pixels get smaller in size, they also become a lot less effective at gathering light resulting in lower sensitivity for the camera. Pixel sensitivity is related to the area of the pixel, so in our comparison, the pixels that are 40% larger will likely deliver twice the sensitivity (1.4×1.4~2). This means that to obtain the same light sensitivity will require an extra f/stop or double the exposure time. Alternatively, the camera’s processor will have to gain-up the signal. (More on this in the next section).
Nikon probably realized that for some photographers, more resolution isn’t always better. Adding pixels means making them smaller and the smaller the pixel the worse the light sensitivity. Fortunately, noise reduction techniques have improved considerably in the past few years, making it possible to increase resolution while keeping sensitivity about constant. So what if you applied current noise reduction strategies to an existing sensor, without adding more pixels? That’s certainly what Nikon intended with the design of the D3S, leaving the resolution at 12.1MP allowed them to increase the sensitivity for those customers that need it. For photographers who absolutely need more resolution, Nikon has the 24.5MP D3X.
For the law enforcement photographer, some of the most demanding situations occur at night when there is little ambient light, or scenes are at a distance. Today, dSLR cameras have unique capabilities and accessories that are available to help the photographer make the most out of night-time imaging circumstances.
Next: Aperture and Exposure Time, Night Vision Modules and more on the following page…