Video surveillance
Like other building automation systems, physical security is migrating from an analog to one that is network-based and IP-focused. That means existing network infrastructures can be used to deliver video security solutions in new and exciting ways. Video surveillance has always to be a balance between the need of security, safety and the demand of privacy. It should be carefully planned, with clear rules and what we expect as a result. A good system doesn’t deliver terabytes of video, which no one could use and control. With time tables and today’s video analytic functions, IP video systems are able just to deliver and record critical events.
There are a few critical parameters we need to understand, to implement a good and reliable video surveillance system.
Field of view: is the first thing, when selecting a camera; That is, the area of coverage and the degree of detail to be viewed. The field of view is determined by the focal length of the lens and the size of the image sensor. It can be classified into three types: Normal View (offering the same field of view as the human eye), Telephoto and Wide angle.
Resolution: The old analog PAL cameras have a resolution of 576 lines. Today’s IP cameras are available from 1,3 to 5,2 Megapixels and more. Depending on the surveillance goal overview, detail and forensic, there a different demands for the pixel resolution per cm at the field of view.
Light Sensitivity minimum illumination refers to the smallest amount of light needed for the camera to produce an image of useable quality. Minimum illumination is presented in lux (lx) and specifies the sensitive of the camera to light. Good IP-cameras have a light sensitivity less 0,05 lux for colour and less than 0,02 lux for B/W.
WDR Wide Dynamic Range is a camera feature, that handles a wide range of lighting conditions in a scene. Dynamic Capture works by acquiring several images with different exposure times for each image. These pictures are then combined in a composite, where both the brightest and the darkest parts are kept, resulting in an image with exceptional clarity and sharpness.
Video Network:
Optimizing the bandwidth of video streams is a vital demand to prevent overload to your network. Besides compression and video analytic there are several methods to optimize the bandwidth and network resources. There are 3 different ways to transport video streams from the camera to the destination:
Broadcast is a One-to-All communication between the source and the targets. This is mainly used in TV-Broadcasting and should not be used for video surveillance.
Unicast is a one to one communication between source and destination and using either TCP or UDP. In this case the camera must be able to establish many connections if more than one destination want to receive the video stream. In this case the bandwidth for four destination is 4 times the bandwidth of the video stream.
Multicast establish no direct connection between the source and the destinations. The connection is established by joining the multicast IP-address group. This allows a significant reduction of bandwidth use. The switch must support in this case IGMP snooping which is supported by most of layer 2-switches. Most of the IP-cameras support this feature for H264 and MPEG-4 but very often not for MJPEG. For good multicast traffic control, the VMS must be able to support following functions:
a) Support of different transport methods in one system
b) integrated proxy-service to convert unicast to multicast and vice versa
c) automatic detection of network resources.
Multistreaming is the ability of an IP camera to deliver 2 or more streams with different quality (resolution and/or frame rate). The second stream could also be used to show only small window of the total scene.
Compression:
technologies are about reducing and removing redundant video data so that a digital video file can be effectively sent over a network and stored on computer disks. The video quality, can be affected if the file size is further lowered by raising the compression level for a given compression technique. There are three main compression standard: JPEG (image compression), MPEG-4 and H264 (video compression) followed by H265 coming in the next years.
Edge storage: works as complement to central storage and allows to record video directly to a SD/SDHC card. It can record locally when network is not available or be used as a pre-alarm buffer to retrieve the last seconds before an alarm triggers video transmission to central storage.
Video analytic:
Is a method for automatic analysing video streams and pictures to detect and determine temporal and spatial events. Video motion detection is one of the simpler forms, where motion is detected in a fixed background. This basic function is implemented in all modern IP-cameras. More advanced functions allow functionalities like video tracking, left or removed items, border line control, licence plate recognition, detection of persons, up to face recognition and detect danger situations like a person has fallen and is lying on the ground.
Video analytic could be performed in the central VMS system or in an intelligent camera. Inteligent cameras allow to design systems, where video transmission to the central VMS system is only performed in case an event has been triggered. As cameras become more and more powerful this will be the future of video surveillance systems.
Integration:
is one of the most important features of a state of the art video surveillance system. Alarm, fire, and access control systems should be able to send triggered event in a fraction of a second to the IP camera to start transmission and recording of the alarm situation. On opposite, intelligent cameras should be able to send triggered events to the alarm system or recognised licence plates to the access control system to decide to open the gate or not. Cameras could also be used just to detect presence to control light and heating or recognize user gestures to control building parameters. The amount of fantastic possibilities for integration is endless.
