Network camera

from Wikipedia, the free encyclopedia
Parking lot surveillance camera

A network camera , also known as the Internet Protocol Camera or IP camera , is a camera for a closed-circuit television system ( CCTV) that provides digital signals at the output in the form of a video stream that is further processed by a computer network using the Internet Protocol IP can be.

Basically every normal webcam is a network camera. In the narrower sense, however, a network camera usually means a surveillance camera in a closed circuit television network. In addition to digital, there are also analog surveillance cameras.

Types of network cameras

There are two types of network cameras:

  • Centralized network cameras are connected to a network video recorder (NVR), which takes over the recording, video processing and alarm management.
  • In addition to the actual camera component, decentralized network cameras also contain a computer that takes on the tasks of the NVR and can transfer the image data to any connected storage medium. The computer essentially consists of a CPU , a flash memory and a DRAM memory. The network camera software makes it possible for the device to act as a web server , FTP server , FTP client and e-mail client in the network.

Differentiation from webcams

Webcams can be seen as the forerunners of modern network cameras. A classic webcam needs a connection to a computer. The data can only be brought into the network from this computer. In contrast, a network camera can independently forward its data to other stations in the network. A webcam cannot be addressed from other stations, but only from the PC to which it is connected. A network camera, on the other hand, can receive all sorts of instructions remotely and, if necessary, can also be supplied with new software remotely. Webcams are simple devices that are mainly used for communication among Internet users. In contrast, network cameras offer a greater variety of functions. For example, they are often equipped with motion sensors and night lenses.


The first centralized network camera was launched by Axis Communications in 1996 under the name Axis Neteye 200. In the beginning, individual web servers were still used to operate the cameras, but from 1999 the company used Linux as the operating system. In 1999 Mobotix presented the first decentralized IP camera, which also contained Linux as the operating system and could independently take over video processing and recording, as well as alarm functions. In 2005 the first camera with video analysis finally came onto the market. The video analysis (Video Content Analytics) was able to determine certain events (theft of previously specified objects, movement within defined limits) and trigger an alarm accordingly.

Advantages over analog video cameras

Network cameras offer clear advantages over analog video cameras, so that they are now well established:

  • The data is already available in digital form so that it can be managed, sent and archived in non-application-neutral IP network structures. This means that operator stations with which live images or recordings from many cameras are processed and managed can be located at remote locations. While analog cameras are usually connected to the associated central technology and monitors via fixed cabling, a network camera requires a connection to an IP network so that its data can be viewed by all authorized network participants. Access to the image material is not tied to a specific device (like a surveillance monitor in analog devices), but can also be done with notebooks or smartphones, for example. Archiving on network recorders also works via an IP network. The Internet can also be part of this IP network.
  • IP cameras can be connected wirelessly to the network and can therefore be set up anywhere within the network range.
  • While an analog video camera requires a local power supply, a network camera can draw power using Power over Ethernet . However, the network must be suitable and equipped for this.
  • The images can be easily provided for many people.

In the past, the decisive disadvantage compared to analog recording devices was that these cameras always provided an unrestricted live image in PAL quality. If a network camera was to deliver images of comparable quality, it had to generate 25 full images per second at 4CIF resolution (DVD quality) and transmit them in the network. In the MPEG-2 process, which was widespread at the time , up to 8 Mbit / second came together. If the video surveillance system consisted of several cameras, restrictions on resolution, compression or frame rate had to be accepted due to the bandwidth of the network. This sometimes resulted in either blurry, rough images or jerky movement sequences. Since the H.264 process became established and ever more powerful networks emerged, the data from cameras with conventional resolutions can usually be processed without problems.

Structure and functionality

Like analog cameras, network cameras work with analog CCD or CMOS sensors. The images are also post-processed in the DSP (digital signal processor) as with analog cameras (highlighting, backlighting, etc.). With the built-in computer, however, the images are then digitized and compressed. Some network cameras have digital input and output contacts. The inputs can be connected to alarm-generating sensors . The camera can react to an alarm in different ways. In this way, images, emails or SMS can be sent. With the outputs, circuits such. E.g .: lighting or hazard alarm systems are implemented. Network cameras often have image memories with a ring buffer function, which can be used to send images that were recorded before the alarm was triggered.

If a network camera z. B. saves images on a 20 GB hard drive in the network, with good compression depending on the frame rate and resolution z. B. save the continuously recorded images of a whole month. However, it is not absolutely necessary for permanent recording. The recording can also be time-controlled or event-controlled (external sensors, motion detection of the camera).

Many network cameras have an integrated microphone and a built-in loudspeaker unit and thus enable bidirectional communication. The microphone can be used for room surveillance, which is why this function is rarely used due to the legal situation in Germany. The loudspeaker can be used to address a person in the area of ​​the camera, but there are only a few applications. In addition to the above, there are other aspects that must be taken into account when installing and using an IP camera - also in the private sector. The secret surveillance of family members or guests is just as forbidden as the surveillance of public roads (e.g. sidewalk in front of the house).


Analog cameras were linked to the PAL or NTSC format (or the resulting 4CIF format) due to their relationship to television . IP cameras, on the other hand, have no restrictions whatsoever when it comes to resolution. In the beginning, VGA resolutions similar to 4CIF were often offered, but today there is an abundance of formats and resolutions up to several megapixels. However, the process of encoding megapixel resolutions in fluid image reproduction in quasi real time is very computationally intensive. It is also in the nature of things that large data streams are generated in this way. Furthermore, megapixel sensors are less sensitive to light than conventional sensors, as the size of the individual pixels is smaller and thus less light per pixel is incident. On the other hand, not every monitoring task requires many megapixels: It is important to choose the right solution for the purpose.

Areas of application

Network cameras are primarily used as surveillance cameras , which means that their use is subject to certain legal requirements (e.g. Section 6b BDSG ). In addition, IP cameras are widely used for commercial applications such as in tourism as a transmitter of live streams of advertised locations, the image documentation of progress on construction sites or the tracking of goods in economic transactions.


In principle, the camera is integrated into an Ethernet structure. Usually copper or (via media conversion) fiber optic structures are used, the connection can also be made wirelessly via WLAN . In practice, DSL or even ISDN are used for an Internet connection that may be required , but mobile radio standards such as UMTS are also conceivable.


As with any network component, there is also a risk of network cameras being accessed by unauthorized persons. Most modern cameras have password protection and similar protection mechanisms that make it difficult or impossible for crackers to take over. In addition, the image data can be transmitted using SSL encryption .

However, some commercially available IP cameras have security flaws that allow unauthorized persons to access the images or the video stream or even to control the camera; in September 2016 IoT devices with outdated was firmware a Linux - malware discovered that over Telnet spread accounts, however.

Individual evidence

  1. a b Central or Decentral
  2. ^ History.  ( Page no longer available , search in web archivesInfo: The link was automatically marked as defective. Please check the link according to the instructions and then remove this notice. Axis Communications@1@ 2Template: Dead Link /  
  3. Mobotix - a niche company is growing seventy-fold
  4. Caroline Machy, Cyril Carincotte and Xavier Desurmont: On the use of Video Content Analysis in ITS: a review from academic to commercial applications
  5. 10 Reasons To Switch From Analog Cameras And DVRs To IP Cameras And NVRs
  6. Video surveillance and web cameras , Independent State Center for Data Protection Schleswig-Holstein (ULD) - October 7, 2015
  7. , February 6, 2012: Back door in IP cameras from TrendNet
  8. , September 11, 2016: Security experts find IoT botnet (September 11, 2016)

See also