With Night Vision Technology
The need to make wildlife observations at night without altering the subject animals' natural behavior has led to the widespread use of night vision technology. However, unless researchers understand how this technology for seeing in the dark actually works, they are likely to overestimate the capabilities of their equipment and the quality of their data.
Image-intensified night vision technology is advertised as giving the ability to "see in the dark". This statement is misleading, as these devices do not see in the dark, but merely amplify existing levels of available light, including light in the near infrared band (750-900 nm). The light these devices detect in this infrared band is additionally converted to a wavelength visible to humans. Let's first consider the amplification process employed by these devices.
The amplification process is not optical, as in the light gathering capability of large aperture binoculars (Fig. 1), but is electronic, more closely related to a stereo system amplifier. With binoculars, the light coming into the instrument from a distant object is magnified by a series of lenses, and presented to the observer's eyes. Note that the light that you see, is the light that came from the distant object, merely magnified.
On the other hand, when light enters a night vision device, it is immediately converted from light energy into an electrical signal (Fig. 2). This electrical signal goes through an amplification process, and is converted back into light for viewing. Note that in this case, the light that you see is not the light that entered the device, but is an electrical representation of the original light.
This distinction is important, because the light-to-electrical conversion, amplification, and electrical-to-light conversion processes have certain "side effects" on what is ultimately viewable with the device. Let's now consider what light is available at night. Figure 3 is a representation of what a good pair of binoculars can do with nighttime ambient light. In the figure, the scale on the left represents a range of light available on a typical cloudless night. The top of the scale represents full moonlight, and the bottom of the scale represents the light level found in deep shadow. The scale on the right is a relative representation of what human eyesight can perceive. What we first notice from this figure is that there is a large range of illumination available at night that our eyes are incapable of detecting. This figure also shows that looking through a pair of binoculars at night does not allow you to see anything dimmer than what you can see without them. The function of the binoculars is to magnify, not amplify.
Now let's replace the binoculars with a light amplifying device, such as night vision goggles. Now we see that this device can take very dim light and amplify it up to the range that we can see. Note that the brighter areas of the nighttime scene, such as the moonlit areas, are too bright for this device, and will produce a saturated image for these areas. Another problem you will see is the introduction of what is called "System Noise" into the viewable image. Because this type of device utilizes an electronic amplification process to see dim light, it suffers the limitations of this amplifier. One of these limitations is the injection of random noise into the image. This noise manifests itself as "graininess" in the image. For conditions where there is sufficient nighttime illumination to use the full brightness range of the device, this noise is a small percentage of the overall image, and is barely noticeable.
However, even with a light amplifying device, there are many instances when there is not sufficient illumination even for one of these devices. This situation is called "light starved", and is shown in Figure 5. In this case, there is insufficient illumination to utilize the full brightness range of the device. Notice that the system noise remains a constant, regardless of the light level, which now makes it a significant percentage of the usable image.
Figure 6 shows two instances of the same image. The one on the left was done with adequate illumination to use the full brightness range of the imaging device. The one on the right is the same image with simulated system noise equivalent to what you will see in a light starved situation.
Now let's consider the case where there is more than enough light for the night vision goggles. This situation will occur on nights with near full moon conditions. This wasthe condition shown in Figure 4. However, this is actually not the behavior of night vision goggles in this situation. Night vision goggles are equipped with automatic gain circuitry. This circuitry has the effect of automatically adjusting the sensitivity of the goggles so that they are only as sensitive as is needed to properly image the brightest region in their field of view. So when bright moonlight is visible within the scene, the goggles will drop their sensitivity to a point where they are not much more sensitive than good binoculars. This situation is diagrammed in Figure 7. In this case, the system noise is no longer a problem, but the device is not capable of imaging any detail in shadowed areas. Animals moving through the moonlit areas will be clearly visible, but any activity occurring in the shadowed areas will not be observable.
In figure 8, the image on the left was done with a device operating at its full sensitivity range. The one on the right is the same image adjusted to simulate excessive brightness. The illuminated regions in the image are still clear, with no noticeable graininess. However, all of the shadowed areas merely appear black.
These limitations of light amplification make it preferable to use night vision devices as wavelength converters of infrared to visible light. Providing supplemental infrared illumination of an appropriate wavelength not only eliminates the variability of available ambient light, but also allows the researcher to illuminate the specific areas of interest. Supplemental illumination can be positioned to eliminate shadows and enhance contrast, and if placed in the same manner for each field observation, improves repeatability.
Although supplemental infrared lighting improves the quality of night vision images, its main value is that it allows the use of solid state cameras, which also have the ability to convert near infrared energy into visible light. They have several advantages over night vision devices. Camcorders possess a higher spatial resolution than night vision goggles as well as a means to permanently record the viewed images. They are readily available and less expensive than good quality night vision goggles.
In summary, camcorders with IR sensitivity such as the Sony Night Shot feature, augmented with adequate infrared illumination, can be used in all applications where night vision goggles are used and produce superior results.