Define the scene and application needs —To determine the type of network cameras required, as well as the number of cameras needed to adequately cover an area, you first need to determine the scene or environment and the goal of the surveillance application. The purpose will determine the placement of the camera, the type of camera and camera features required (e.g. progressive scan, megapixel for exceptional details, audio, security features) and lens adjustment/type (normal, telephoto or wide angle). A security operator using a PTZ or dome camera can cover a large area and capture different images for different purposes.
In many cases, different cameras will be needed to capture images for different purposes (i.e. one camera providing a full overview image for capturing an incident in action, and another camera for close-up views of a person/object for identification purposes).

Consider —

• Environment. This will determine whether you need an outdoor or indoor camera, whether the camera needs to be tamper or vandal proof, and whether special housing is required.
• There are network cameras for indoor use, as well as ones for both indoor/outdoor conditions;
• Indoor/outdoor cameras have varifocal lens that automatically adjust the lens’ iris;
• Also available day/night cameras, which provide color images during daytime and black & white images during nighttime.
• Lighting requirements. Is there adequate light to obtain a good quality image? Do you need to add light sources? How light sensitive should the camera be?
• At least 200 lux is needed to capture good quality images.
• Area of coverage — wide, narrow, general or detailed. Angle of view needed.
  Determine how much of the scene you need to see. Network cameras come with fixed angle and focus, as well as variable ones that allow remote pan/tilt/zoom capability, which enables a wider area of coverage. The bigger the area, the more cameras are needed.
• A PTZ or dome camera is able to cover a wider area than a fixed network camera.
• Distance from position of camera to object being monitored. This determines the type of camera and type of lens (normal, telephoto, wide-angle) to use, as well as the placement of the camera(s).
• Certain network cameras have lens that are replaceable.
• Determine the kind of surveillance you want to conduct — overt / covert. This will help you in selecting cameras that offer a non-discreet or discreet installation.
• Determine the kind of image you want to capture — overview or close-up for identification purposes.
• High or low traffic.The higher the traffic, perhaps the more cameras are needed.
• Viewing and recording needs. Determine when and how often you need to view and record: day, night and/or weekends? This will determine the features you need to consider; i.e. frame rate capabilities, type of video compression, bandwidth saving features such as video motion detection, and alarm management functions.

Not all network cameras are created equal. A key determinant of a network camera is image quality. When assessing image quality, be sure to consider a network camera’s light sensitivity, the crispness of moving objects and the level of clarity. Read through a camera’s datasheet and, most importantly, field test a few cameras before making a decision. A network camera with an open, application programming interface enables a large variety of software vendors to write programs for the cameras. This will increase your choices in software applications and will ensure that you are not tied to a single vendor. Find out if the network camera is able to work with the software of your choice. Your choice of network camera should never limit vendor options and functionalities.

Determine your application purpose — simple remote viewing or intelligent surveillance system with advanced event management, input/output triggers, audio component? Determine when and how often you need to view and record — day, night and/or weekends? Schedule the needs for every scene.

• Different scenes may require different monitoring schedule, use of triggers and different viewing and recording frame rate. Be realistic about when you want to view and record and at how many frames per second. Keep in mind that you will probably not want to record uninteresting video 24 hours a day. You can schedule the recording based on schedule, video motion detection or alarm triggers.
• If you want to simply capture the identity of a person at an entrance, you do not need the camera to continually send video. You can set it up so that only when the door opens, is the camera triggered to capture and send the necessary image frames. However, if you want to monitor behavior patterns, you may want to view and record at a higher frame rate.
• Determining your monitoring schedule will also help you assess the level of network use and whether you'll have to install additional bandwidth capacity on your network or whether you can make use of the same network as for general business activities. Network traffic is generally lower during nighttimes and weekends, which may coincide with when you want the network cameras to operate. Hence, you can take advantage of the already available bandwidth without any additional network installation.

Full frame rate on all cameras at all times is more than what is required for most applications. With the configuration capabilities and built-in intelligence of the network camera/video server and/or video management software, frame rates under normal conditions can be set lower, e.g. 1 to 3 fps, to dramatically decrease bandwidth consumption. In the event of an alarm, i.e. if motion detection is triggered, the recording frame rate speed can be automatically increased to a higher frame rate. In order to appropriately calculate the storage requirements of a network surveillance system, there are a number of elements to factor in.

1. Image size x frames per second x 3600s = KB per hour / 1,000 = MB per hour
2. MB per 24h x requested period of storage = Storage need

Network video products utilize network bandwidth based on their configuration. Bandwidth usage depends on five criteria: image resolution, compression type, compression ratio, frame rate and image complexity:

• Image size x # of frames per second x 8 Kbit/s = bandwidth use of one camera in Kbit/s.

The above criteria can be set either in a video management software or in the network camera or video encoder product itself. A simulation-based calculators helps to provide guidance on a network video product’s bandwidth and storage requirements.

Assess network use of current LAN. What are you or the company using it for? There are software and hardware tools available to measure the level of congestion on a LAN. Talk to your company’s network administrator. Assess network use of current WAN links. WANs are geographically dispersed networks. You may have WANs that span between office buildings. You can measure the level of network congestion.
Determine the pattern of congestion levels over a given periods — It may be that the network traffic drops off during nighttime and weekends. The usage pattern will help you to determine whether you can:

• simply use existing network infrastructure as for general purpose network needs (i.e. same cabling and outlets, same switches if there is sufficient capacity) to add network video surveillance equipment;
• use a combination of existing general purpose network as well as new network for IP-Surveillance;
  — or —
• use a separate network altogether for IP-Surveillance purposes.

Do you need to add new equipment to the network? If additional network capacity is needed after assessing your existing network(s), new cabling is normally not needed. You may simply need to add a switch or reconfigure the patch panel. The network provider or administrator, usually the IT department, will have a set of security policies in place for network usage. These policies include items like —

• log-on credentials,
• back-up procedures,
• and virus filtering and scanning.

Many of these policies can affect system performance. For example, are external connections to non-corporate machines allowed? Such connections will be needed if the organization plans to use external alarm monitoring services. This would raise a host of questions —

• Will this alarm monitoring center have the capability to connect to the local site to do a guard tour, either locally or through a remote alarm company?
• Will video be stored, and in what manner?
• Should these stored images be included in general back-up procedures?
• Is the current back-up system able to handle the additional data?

These are just a few examples of questions and policies that would need to be explored to assess how network security procedures can impact system performance.