In many cases, picture quality is improved with DVRs, their images can be handled and stored for long periods of time without image degradation, and tape breakage is a thing of the past. As one manufacturer compared it, the effect of DVR usage is similar to the improvement from typewriter to word processor.

Video events now can be sent to and viewed from anywhere in the world and even accessed wirelessly. DVR technology allows instant access to events by date, location and sometimes even by the type of incident or person involved. A DVD or CD of an event can be downloaded and handed to authorities in a matter of minutes with DVRs.

DVRs also have increased the storage capacity of recorded events. They have allowed a greater variety of types of recording than ever was possible with videotape. For example, 30 fps video can be displayed live on monitors that guards watch while 15 fps is being stored to reduce storage needs. Frame rate and picture quality can be increased when high-interest events are triggered, and low quality used when static events are occurring in camera range. Some DVRs allow for notification by e-mail of alarm events. Such versatility was only a dream a few years ago when DVRs were just in development.

The real revolution has been in affordability. The predictability and reliability of DVR operation is bringing new customers to their use. Higher levels of security are available in light commercial applications that could not justify the expense of video surveillance before now. Even private residences use DVRs for surveillance.

Tips for a Successful DVR Installation

• Make sure the DVR selected truly meets a customer’s requirements and is fully understood. Check closely with the customer about his or her expectations to ensure they match the equipment’s capabilities.
• Security dealers and systems integrators should provide training and support for customers, and manufacturers should do the same for dealers and integrators.
• Provide different levels of training for the appropriate users at the customer’s site. Some users only will need to be trained in operating day-to-day features, while others will need higher-level administrative training.
• Provide training material for features that are used infrequently.
• Be sure to read the instruction manual before installing DVRs. Although seemingly obvious, that is one of the most important things that can be done to facilitate proper installation and is done surprisingly seldom, manufacturers maintain.
• One manufacturer’s technical support department receives the most telephone calls at the end of each day as installers realize they do not know how to operate the DVR and are rushing to complete a job after stringing cable all day.
• It is important to pay attention to the mean time between failure of DVRs. A DVR with a failure rate of 50 percent would lead to a great loss of time and money for installers, manufacturers and end users.
• Make sure the DVR is locked up for security, but kept in a ventilated place where it can be cooled.
• When possible, stage equipment before going on-site to assure its accurate operation.
• Finally, manufacturers recommend choosing a vendor that offers stability in the industry. Many DVR manufacturers come and go, they point out.

Connecting a DVR

Many use RG-59 coaxial cable for connections and have a BNC connector, which is the main input/output of a DVR. Video monitors also use BNC connectors.

Composite outputs for red/green/blue elements of a picture may be included on DVRs, along with S-video ports or VGA ports, which are similar to those used with computer monitors.

DVRs are made to connect with analog cameras, although some include connections for Internet protocol (IP) cameras. Analog cameras can be connected with BNC coaxial cable or with an optional TP-to-BNC converter. Often IP cameras are connected to different devices such as hybrid recorders or network video recorders.

Other types of cable that can be used with networked DVRs include Cat 5e, which is Ethernet cable, with RF-45 connectors or twisted pair. Fiber-optic connections and even encoders to transmit camera signals over an IP connection can be used.

Networked DVRs with Ethernet connections can be accessed from anywhere in the world.

An Ethernet port is used for an IP connection or coaxial cable. For a serial connection, RS-485 or RS-422 is used.

Special Functions & Features

This includes intelligent video, such as when something moves outside a box that is “drawn” on the screen. With pixel search, a portion of the viewing area can be identified and whether movement occurred there can be determined.

An intelligent or pattern search seeks an image for a patterned time frame, such as once at the beginning of every hour in a 24-hour time period. Advancing to the beginning of each 10-minute period and then to the beginning of each minute will allow investigation of the beginning and end of an event.

Video content analysis (VCA) is improving all the time and recent advances have allowed for more accurate detection of certain events like trip-wires, idle or loitering people, cars parked in no-parking areas, left-behind objects and other events.

DVRs are starting to incorporate this technology, which previously only was available from a third-party device that the DVR plugged into or as part of a networked system.

Most big-name video surveillance manufacturers will have this capability built into some high-end DVRs in the next year or two. However, manufacturers warn that VCA’s capabilities sometimes are oversold beyond what actual products can deliver today. They recommend seeing actual demonstrations of VCA capabilities before purchase.

Test the search capabilities of DVRs you are considering to see how intuitive they are.

Picture Quality

Wavelet is another type of compression that is faster to transmit. Other types include JPEG 2000, regular JPEG and regular MPEG. The type of compression is included with the DVR.

Other types of compression, such as MPEG-4, can be used to conserve bandwidth and storage space. These use sophisticated algorithms to reduce storage requirements. For example, when a portion of a picture does not change, the information for it does not have to be repeated for each frame of the video. It is simply held there until something changes in the picture. This can be valuable in surveillance, where views frequently do not change.

One possible solution for the bandwidth storage problem is called transcoding. It allows recording in one format such as JPG for storage in a server, but can stream over a network a more compressed MPEG-4 or wavelet version that has a lower bit rate. MPEG-4 can use one-fourth or less of the bandwidth that JPEG does.

MPEG-4 will yield greater efficiency than Wavelet at frames rates higher than 2 images per second (ips) because the average file size is constant with Wavelet but variable with MPEG-4. A file size of 7 KB with Wavelet would stay at 7KB no matter what the frame rate is, but with MPEG-4, the average file size could drop to 1KB at higher frame rates because of the codec’s additional compression, manufacturers calculate.

The capacity of the DVR’s processor determines how much video can be compressed. So unless a processor is large enough, it may not be able to compress high frame rates at high quality.

Many DVR manufacturers will be moving toward the ability in the next six months of allowing each user to select how to view the video remotely. Regardless of how the video is viewed remotely, the actual recording that is archived is never changed. Viewers can switch from streamed MPEG-4 to high-quality JPEG at key times when deciding whether to take action.

Calculators are available or on manufacturers’ Web sites that determine the sizes of files with certain compressions methods. The size of a hard drive determines the amount of picture quality and frame rate that can be stored. That is why selecting a DVR with a high hard drive capacity or one that can be upgraded to a high-capacity hard drive is important.

Ultimately, the relation between picture quality, frame rate and storage capacity is a trade-off based on the exact use of the DVR and what is being recorded. An installer’s goal is to not allow a DVR to start overwriting previously recorded material until the client has obtained everything needed from the DVR.

This article is based on interviews with Justin Lott, product marketing manager, Bosch Security Systems, Fairport, N.Y.; Don Taylor, vice president of marketing, and Chris Villareal, product marketing application specialist, of Dedicated Micros Inc., Chantilly, Va.; Jordan Shishmanov, product manager, Honeywell Video Systems, Louisville, Ky.; Vy Hoang, executive vice president of sales and marketing international, i3DVR International Inc., Scarborough, Ontario, Canada; Yvonne Schwemmer, product sales manager, Pelco, Clovis, Calif.; and Gary Perlin, vice president of video products of Speco Technologies, Amityville, N.Y.

side bar: Features You Will Want on Your Customer’s DVR

• Ethernet connections so a DVR can be accessed from the Internet are becoming standard on some models. So are built-in multiplexers.
• Internal or external archiving on CD-ROM, DVD, USB stick, thumb (flash) drive, or CF card also is an important feature to have.
• Event recording, which is the capability to save hard drive space and record only when an event occurs, is a necessity.
• DVRs should record by schedule recording at a certain date and time, when alarms are detected or certain types of motion detection, either simple (when any motion is detected) or sophisticated (when certain motions in certain areas, such as a boxed-in area, are detected, or when an object is moved or does not move).
• Besides global settings that apply to all cameras connected to a DVR, the capability for individual setting of each camera should be included, so one camera can record at 30 images per second (ips), for example, while another records at only 2 ips. Each camera also should be programmable in terms of camera scan patterns and resolution, if applicable.
• Selecting different resolutions can aid storage capacity. A good low-resolution size is 320 pixels by 240 pixels. Good resolution should be 640 pixels by 480 pixels, and the best resolution would be 720 pixels by 480 pixels, some manufacturers specify.
• Specifying a particular type of hard drive is important. Make sure a large enough hard drive can be included to provide customers with the image quality and frame rate they want.
• Flexible storage capacity and the ability to add capacity are important. Being able to remove hard drives when they are full and insert new ones is an inexpensive way to increase capacity.
• At least 30 images per second on one or more cameras and the ability to connect from eight to 16 cameras depending on the installation’s size are important.
• Triplex operation that allows recording, viewing and playback at the same time is a versatile feature. Networking so camera output can be viewed on a local or wide area network or the Internet also may be needed in certain applications.
• Full and multi-screen preview with a graphical user interface is helpful.
• Alarm inputs and audio inputs are handy, as is speed dome control, which allows control of pan-tilt-zoom (PTZ) cameras.
• Easy set-up is a feature of many DVRs now. People should be able to understand how to use it and it should be as simple as possible to operate. But it also should have restrictions on who can access its administration.
• Some method of approved watermarking should be available with the DVR to establish a recording’s authenticity in case it is used by police or in court.
• When selecting a DVR, pay close attention to the specifications in the data sheets. Watch out for settings that may not fit an application. For example, one specification for a DVR may only apply to playback, not recording, or high frame rates may only be available at lower resolutions.