High-speed PTZ color IP video cameras.
Four definitions with one noun. Separately, PTZ cameras (including high-speed ones), color cameras, and IP cameras have long been familiar.
Briefly technical and tactical features of each category.
The general name IP video cameras refers to any cameras that can be directly connected to computer networks to broadcast a video signal.
Now this definition includes both the simplest devices capable of transmitting 1–2 frames per second and powerful systems with adaptive coding in real time, simultaneous image analysis and flexible two-way communication with the observation post.
The most important parameter of all cameras with digital (colloquially – IP) video signal transmission is image quality. There are no objective criteria for coding quality.
Of course, such obvious parameters as the number of frames per second and the number of pixels on the screen reflect quality. However, most IP cameras today declare 50 frames per second (apparently 25 frames — a regular video signal has 50 half frames per second) and a D1 screen size (usually 640 x 480) or CIF (320 x 240). Formally, analog video recorders provided quality comparable to the CIF screen size, so this is enough. But in fact, after digital processing and compression, the image may not have a horizontal resolution of 240 TVL, although formally 320 pixels should have provided this. In addition, it may turn out that the camera transmits 25 frames per second only when there is no movement in the frame, and the more movement, the fewer frames per second it can process.
Finally, another parameter of IP video cameras, significantly antagonistic to the previous one, is the volume of the digital data stream they generate. The first cameras took 20-30 megabits per second. That's a lot. Modern cameras (as advertised) usually have 50-200 kilobits per second. Don't think that this is necessarily good. The smaller the data flow, the less information is transmitted.
Miracles do not happen.
If a modern camera capable of compressing a video signal to 50 kilobits/second has an operating mode of 20 megabits/second, in this mode the image quality is probably really good. But in the low-band mode (50 kilobits), the quality certainly leaves much to be desired.
Pay attention to one more parameter: can the encoder provide a fixed data flow or can the data flow increase sharply if there are difficulties?
In the second case, your network can become swamped when all cameras simultaneously increase their data volume for a short period of time, such as when a gust of wind causes trees in front of all cameras to move at the same time.
A significant advantage of IP video cameras is the ability to transmit non-standard frames. While the security industry was tied to a standard analog video signal, it was impossible to even think about megapixel cameras or high-speed 100-frame devices. By removing the limitations of the old standard, IP video cameras expand the options.
In addition to video cameras with a built-in encoder (they immediately have an Ethernet connector or a Wi-Fi antenna), conventional cameras with an external IP encoder are often used. This solution provides a greater choice of video cameras and lenses, but an external encoder obviously does not allow you to control the parameters of the video camera itself (which is sometimes useful) and is not adapted for a specific task (for example, for processing the movement of a PTZ camera). In addition, an external encoder means the presence of an unnecessary double conversion digital — analog — digital. After all, the signal taken from the CCD matrix is already digital to some extent, and if the camera has built-in digital processing, then there will obviously be a conversion to digital, back to analog, and in the encoder again to digital. Note that the above applies only to the case when the IP encoder is an integral part of the video camera.
Often, a seemingly «integrated» camera with a built-in IP encoder actually contains a regular camera and a separate encoder in a common housing. In some cases, this means that the encoder is optimally matched to the camera, but more often than not, if you buy a separate camera and a separate encoder, included in the kit, and connect them yourself, you will get the same thing, but you will have the opportunity to choose where to mount the encoder. And the total cost will most likely be less — they charge a significant amount for pre-assembly at the factory.
PTZ cameras, as a rule, in addition to rotation, are equipped with a zoom lens, allowing you to change the angle of the field of view (zoom in/out of the object). The main advantage is the ability to carefully examine the object of interest and track its movement. The only drawback (in addition to complexity and price): in the absence of a special auto-return function, PTZ cameras are often forgotten in a position when the screen shows a close-up of a section of the territory where something happened last year, and the rest of the territory that this camera should monitor is not visible. Another drawback is that at night it is very difficult to provide illumination only for the camera's field of view; you have to illuminate the entire territory. The main parameters of PTZ cameras are rotation speed, permissible rotation angles (not all cameras are able to rotate in one direction all the time, in some cases the wires wind around the base, and the camera stops), the range of adjustment of the focal length of the lens.
When applied to IP video cameras, PTZ cameras have another significant drawback: when the camera rotates, the image changes rapidly across the entire screen. This is a serious test for all compression algorithms (if they are even capable of handling such a signal). Some specialized PTZ IP video cameras are said to have special compression algorithm properties that allow them to transmit the signal well while rotating. However, these properties are either completely non-standard or boil down to the fact that the encoder safely switches to low-frame transmission mode without interframe compression. However, most encoders will not transmit anything at all until the camera stops and the image stabilizes.
Modern high-speed cameras almost always have a design in the form of a spherical cap (hence their second name — dome cameras) and a balanced module camera + lens placed inside it on a gimbal suspension. Such a design with an extremely lightweight moving module allows you to achieve a rotation speed of 360 degrees per second or even faster. Before the advent of dome cameras, there were rotating devices with the same speed, but they were heavy and therefore required a reliable foundation, almost like for a milling machine. Modern dome high-speed cameras can be mounted without thinking like ordinary ones, on lightweight compact brackets. The disadvantage of dome cameras is the impossibility of using windshield wipers, as well as reduced thermal insulation. However, there are options for mounting such a camera in an additional dome, a kind of double-glazed window is obtained.Please note: not all PTZ cameras are dome cameras, and most importantly, not all dome cameras are PTZ cameras. There are spherical (hemispherical) housings for fixed cameras. Cameras can also be rotated there, but only manually during setup. Such housings make it easier to adjust the camera and hide the camera's direction of view from outsiders. There are housings for any standard camera, and there are also so-called mini-Dome housings with a frameless camera inside the cap. Models designed for ceiling mounting cannot always be placed on a wall — when rotating, the coordinates are twisted, the camera will turn around its axis relative to the horizon, and the image on the screen will be tilted.
Color cameras, unlike black and white ones, provide a more natural, familiar to the eye image, and more informative. However (again, in addition to the price) they have one significant drawback — reduced sensitivity. In addition, they practically «do not see» in the infrared range, and therefore at night they necessarily require natural white backlighting. Finally, the standard for transmitting a composite color video signal (the so-called low-frequency video signal) means that the system resolution will be greatly reduced — to 300-330 TVL. Theoretically, the standard itself describes ways to preserve the resolution, but this requires too expensive circuit solutions, which is unacceptable in security video surveillance.
Many color cameras today have a «night» mode, in which they produce a black and white signal. Most cameras simply turn off the color signal, but nothing else changes: neither the resolution nor the sensitivity, only the unpleasant color noise disappears. The gain in sensitivity is only 1.5-2 times. In some cameras, when switching to night mode, the infrared filter, which is mandatory for color cameras, is mechanically removed, and its spectral characteristic is significantly expanded, which increases the sensitivity several times more. However, mechanically moving the filter is a complex and expensive matter.
When applied to IP video signal transmission, color doubles the amount of information and tightens the requirements for the encoder. Very often, after the encoder, a black and white image seems acceptable, but in a color image, streaks (artifacts) appear on sharp color boundaries or, conversely, on smooth gradients. In addition, if the video camera has the ability to switch to night mode, then built-in encoders are preferable, which will allow you to manually remotely activate this mode, rather than relying on automation. However, PTZ cameras should always be controlled, and if the IP encoder is able to control it, then, probably, it will be able to give commands to switch modes. The main thing is that it is fully compatible with this camera — preferably, from the same manufacturer.