The KODOS video surveillance system. History of development.
Purpose of the system
The KODOS-VIDEO NETWORK system is designed for use on objects of various sizes.
The system supports the following main functions:
- Digitalization and compression (multiple reduction of the space occupied by each video frame) of the video signal received from video cameras
- Displaying the video image on the screen of a local or remote (network-accessible) computer
- Archiving the compressed video signal, both on a local and on a remote computer, with the ability to remotely view the video archive over the network.
- Software motion detector with the ability to edit video frame zones and adjust the sensitivity in each of them.
System composition
The system is based on client-server architecture and supports both one (in stand-alone mode) and any number of computers connected to a network.
Video server – a computer with connected video cameras.
Transmitter – a process running on the video server and transmitting video over the network. Receiver – a process running on the client machine (generally, on any computer on the network), receiving video over the network.
Stand-alone mode
In standalone mode, the system operates under the control of one computer (video server), to which a video camera or other source of analog video signal is connected via a video input board or USB bus.
Up to 4 such boards can be connected to one computer.
Video cameras with analog video output, video multiplexers, quads, matrix switches and other video devices can serve as a source of video signal.
The operator controls the zoom and pan and tilt of the video cameras using only the mouse and keyboard. All actions can also be performed automatically — by timer, motion detector, etc.
Network operation
In network mode, all functions of the standalone mode are supported.
In this case, the operator, located at the client computer, controls the transmitter in the same way as if he were working directly at the video server. One or more receivers can be connected to the transmitter.
The network can have an arbitrary number of transmitters and receivers (limited only by the network's bandwidth), and a connection to each of them, if access rights are available, can be made from any computer in the network.
Network traffic management
The receiver sends requests to connect and disconnect from the transmitter. When several receivers are connected to one transmitter, the network traffic increases by the corresponding number of times. When all receivers are disconnected from the transmitter, the transmitter stops transmitting and does not load the network.
Remote video recording
Recording (simultaneously with video transmission over the network) can be done on both the client and server computers.
Viewing the video archive
The system supports both local and remote viewing of the video archive, with the ability to quickly view, view back, and frame-by-frame playback.
Also supported are archive cleaning and defragmentation, printing and saving individual video frames in JPEG format.
Hardware requirements
Pentium MMX processor and higher (outdated)
A higher processor clock speed will allow recording with a higher number of frames per second, all other things being equal.
RAM 32 MB
Video input card with support for Video for Windows format with 24-bit color representation.
Video camera — Analog video output
Network — Local area network, modem line, radio channel, Internet Network bandwidth 10 Mbit/s, 100 Mbit/s.
Higher network bandwidth is preferable, as it will allow video signal transmission at a higher speed, all other things being equal.
Software requirements (outdated)
Operating system Windows 95 OSR2 or Windows 98 Drivers Direct-X 7, Intel Indeo 5.1
Video signal compression
A special video compression algorithm based on the latest wavelet video compression technology allows to significantly reduce the disk space occupied by each frame and relieve the computer network.
Unlike similar algorithms, wavelet provides high quality video signal even at maximum compression levels.
Video compression parameters
The operator configures the following video compression parameters: compression quality compression algorithm video digitization speed
Compression quality
Compression quality is changed by the user from 1% to 100%. The higher the quality, the more disk space each video frame takes. The figures below show two video frames saved with 100% and 1% 100% quality.
Compression algorithm
The compression algorithm is selected by the user from two fixed values: «fast» («quick compression») and «slow» algorithm. Thus, on a Pentium-200 computer, the speed of 15 frames/sec is supported by the «fast» algorithm, and 3 frames/sec by the «slow» algorithm. algorithm, all other things being equal.
Video sampling speed
The user can set the video sampling speed (number of frames per second) from a fixed series of values: 1, 2, 3, 4, 5, 6, 8 and 15 frames per second.
In addition, it is possible to select any speed value that is 1/N, where N is a positive integer other than 0.
In this case, video digitization will be performed at a rate of 1 frame per N seconds. The lower the rate, the lower the system's requirements for the processor, disk memory, and network, all other things being equal.
Automation of system reactions
The system supports automation of actions (turning on and off video recording, changing the video digitization rate, controlling video camera actuators) depending on external events.
The following information is used as initial information: current time and date information coming from security and fire sensors information coming from the ACS «KODOS» (entrances and exits of users through doors and turnstiles of the protected facility) information obtained by software analysis of video images (motion detector).
Accounting for repetitions
Each frame is compressed separately from the others. On the one hand, this somewhat reduces the overall compression level, and on the other hand, it allows, in case of excessive processor or network load, to skip individual frames without deteriorating the quality of the rest.
Motion detector
The software motion detector is an algorithm that detects changes in a video frame (the number of changed facets — 4*4 pixel blocks — and the intensity of their change) between the previous and current frame.
The main parameters of the motion detector:
Sensitivity — an abstract parameter that characterizes how much a facet must change for its change to be recorded.
The higher this parameter, the more the facet must change.
Trigger Limit — how many simultaneous triggers (changes in the image) must occur for the detector to trigger.
Indirectly characterizes how large an object must be for the detector to react to its movement.
Since the frame may contain constantly moving objects (for example, tree foliage, people walking past the window, etc.), the system has the ability to select frame zones (this is done manually when setting up the system) and analyze each of these zones separately using the motion detector.
Examples of using the motion detector include turning off or slowing down recording if there is no movement in the frame or in specially selected parts of it (thus significantly saving hard disk resources by eliminating the recording of useless images), turning the camera «in the direction of movement» when «a moving object gets into the frame.»
Sound support
The system allows you to record sound synchronously with video. Requirements — DirectSound 7 Characteristics approximately correspond to mp3 22 kHz Mono
Black list
Motherboard GA-BX2000+ 440BX/GIGABYTE Reason — non-standard Promise UDMA66 controller, installed on one interrupt with any PCI port + in BIOS you cannot assign ports to boards.
Update for 2023. (to be supplemented)