The latest and greatest body cam debut got me to thinking about the evolution of these devices. There are relatively inexpensive models you can buy for $200 or so, and some officers are getting these for themselves as a safeguard against accusations of misconduct. Are these good enough to do what cops need them to do?
A standard video image contains a bit over 300,000 pixels (the little dots that make up the picture), arranged in a matrix 640 pixels wide by 480 high. This is the image we lived with on most television sets until flat screens the size of billboards came along, and it’s what most dash-cam video devices continue to produce. It usually works well enough for a vehicle-mounted camera and recorder. The distinction is that body cams are, by definition, not vehicle-mounted. If the camera is attached to your head via a pair of wraparound eyeglasses or some other headgear, it’s going to move laterally a lot faster than your dash cam is likely to do, unless you drive sideways a lot (I had a partner who did that). It’s also going to move vertically as you look up or down.
We don’t notice a lot of movement in our field of vision when we look at a fixed object while turning our heads. This is because our eyes and brains have this cool tracking feature that moves the eyes in one direction while the head turns in the other. The benefit of this becomes very clear if you look at a video clip where the camera is panned too rapidly. Everything becomes a blur, because that video image is really 30 still images recorded every second. Set your still camera for a shutter speed of 1/30, and most of your pictures will come out blurry. Things can move quite a bit in 1/30 second.
640x480 is enough resolution when something is relatively big in the frame, but it breaks up pretty fast if you need to zoom in. We’ve all watched dash cam footage of struggles during a traffic stop, and it’s usually very difficult to tell who is doing what at any given second.
My impression is that there is still room for improvement with these devices, which don’t fully exploit the available technology, to wit:
• Boosting the frame rate and image resolution would do a lot to address the blurring and detail problem, although the file size of the recordings would increase dramatically. There is already a glasses-mounted camera and recorder that will produce 720p resolution (1280x720 pixels—three times that of standard video) at 60 frames per second. I want one, although I have no clue what I would do with it.
• An hour of standard video comes in at about 1GB (your mileage may vary). The gadget described above would produce about 6GB per hour. However, 128GB memory cards smaller than postage stamps are now available for less than $200, and that’s enough capacity for several high-definition Hollywood movies.
• Battery life. Power is a problem for devices like this, which pull quite a bit of juice when they’re recording. You might have enough storage capacity for 12 hours of continuous recording, but not nearly enough power. This is an area where police tech vendors need to establish industry standards that would permit multiple devices to run from a single battery or fuel cell, probably in a carrier integrated with a load-bearing vest or body armor shell.
• That could mean consolidation of multiple devices and elimination of multiple cables, especially those running to the officer’s head. Build the camera, a radio earphone jack, and a microphone into the eyewear frame, and eliminate all but a single cable running to the control module and battery. In theory, this could also include a head-mounted display that could be available soon.
Some of this wish list sounds like science fiction, but it’s all do-able with presently available, off-the-shelf technology. Much of the work lies with getting the police equipment industry to agree on some standards that make equipment more interoperable. From the perspective of someone who started their police career when many cops didn’t even have portable radios, it’s all amazing stuff.
