Video Surveillance Standards, Guidelines, Best Practices

Tim Sutton CPP, PCI, PSP, CHPA

The question of the effectiveness of live monitoring of video feeds has been asked for as long as the
practice has existed. Images of an officer sleeping in front of a desk full of monitors with camera feeds
or eating a donut and reading a magazine while watching the monitors for intruders or illicit activity
have been used as comedic references for years. These derogatory accounts have helped feed the
misconception that a security officer with a duty to monitor live video surveillance is a waste of
resources. I once had a CEO ask me, “why am I paying $15/hr for a guy to watch TV?”

According to a U. S. Department of Justice Office of Justice Programs National Institute of Justice

publication titled the Appropriate and Effective Use of Security Technologies in U.S. Schools A Guide
for Schools and Law Enforcement Agencies from September 1999, experiments were run in the late
1970s for the U.S. Department of Energy to test the effectiveness of an individual whose task was to sit
in front of a video monitor(s) for several hours a day and watch for particular events. These studies
demonstrated that such a task, even when assigned to a person who is dedicated and well-intentioned,
will not support an effective security system. After only 20 minutes of watching and evaluating monitor
screens, the attention of most individuals has degenerated to well below acceptable levels. Monitoring
video screens is both boring and mesmerizing. There is no intellectually engaging stimuli, such as when
watching a television program.1

The practical upper goal that is most frequently established is 16 video monitors per operator. However,
the actual number at which performance (detection and interpretation capability) becomes
unacceptable depends upon many factors, including: (1) how crowded the scene is and how much
activity is taking place; (2) whether or not multiple cameras are sequentially scanned on individual video
monitors; (3) how far into the work shift the operator is; (4) what other duties the operator must
perform; and (5) operator experience.2

Of over a dozen existing CCTV security systems examined, the number of monitors in the control center
ranged from 2 to 55, and the number of operators ranged from 1 to 3. In some of these systems, the
number of operators varied during the day as a function of predictable peak activity levels. As a
representative example, the second largest system examined has 38 monitor screens (no sequencing)
and employs 2 civilian operators most of the time, going up to 3 operators during rush hours in the
morning (7:00-8:45 AM) and afternoon (4:00- 6:30 PM), and dropping to 1 operator after 10:00 PM. This
particular system is "shut down" from 12:30-5:30 AM, but a police duty officer is present and can
observe the screens during that period. The operator’s responsibilities, other than the video monitors,
include patron assistance telephones, police telephone and radio, remotely controlled access to rest
rooms, intrusion alarms, elevator telephones and the public address system. Sequential switching of 2 to
3 cameras per monitor screen was tried and rejected when it was found to degrade operator attention.
To help reduce fatigue, operators receive a 10-minute break about every 1 to 1-1/2 hours. Relatively few

1
https://www.ncjrs.gov/school/state.html
2
Dauber, R.L., "Guidebook for the Provisions of Passenger Safety and Convenience in Automated Guideway
Transit: Final Report," Vought Corp., Dallas, TX, under subcontract to Dunlap and Associates, Inc., Darien, CT, for
Department of Transportation, Transportation Systems Center, Cambridge, MA, Report No. UMTA-MA-06-0048-
78-8 (January 1979). Available: NTIS.
researchers have conducted relevant scientific studies on the number of video monitors one operator
can view effectively. Target numerosity studies on air traffic control type displays have demonstrated
the degradation of detection and recognition as the number of significant targets, or even the number
of irrelevant forms, increases. More applicable studies involving outdoor television scenes of people and
vehicles showed that police officers were better than civilians at detecting critical incidents; near
incidents were detected more reliably than distant incidents; and learning what to look for masked
fatigue effects (even after 4 hours).3, 4 , 5 , 6

The most relevant research studies involved the problems of monitoring numerous video displays for
one or two hours. In one study, 16 and 24 television displays showing little movement were observed. In
the other, 16 displays showing a great deal of movement were observed. Both required the detection of
incidents in prison environments, by civilians and police officers. The video display monitors were 40 x
30 mm (about 19 inches diagonal) in size, and typically arranged in a 4 x 4 or wider 4x6 array at a
distance of 13 feet from the observer. In some parts of the tests only 1, 4, 8 or 9 of the video monitors
were used, but the array was maintained. It was found that, for scenes involving a great deal of
movement, a 100% likelihood of detecting incidents requires viewing only one monitor display. If it is
necessary to watch more than 1 screen with a great deal of movement, then no more than 9 should be
viewed by one person. The likelihood of detecting incidents on 9 monitors was 83%, and on 4 monitors
it was 84%—no significant improvement by going from 9 to 4 monitor screens. However, for 16
monitors that figure dropped to 64%. If very little movement is present in the viewed scenes, then 16
monitors is the recommended maximum number by the authors. When 24 monitors with very little
movement were viewed, the likelihood of detection dropped from 97% (for the 16) to 88% (for the 24).
It is recommended that these same guidelines of a maximum of 9 or 16 monitor screens per operator be
followed by transit systems as well, for the high and low movement conditions, respectively.7 ,8

Where security of busy transit stations is the primary mission, eyes should be kept on the screens
virtually full time; non-viewing activities increase the probability of missing the detection of a brief
critical incident and lengthen reaction time generally. In one large department store examined,
detective operators only attend the monitor screens for just one hour at a time, rotating duties with
other roving store detectives throughout the full workday. The tendency among existing CCTV transit
security organizations is to maintain the conventional, non-rotating work shift of about eight hours'
duration. However, these organizations recognize the need for breaks from the routine of watching
monitor screens. Such breaks are provided in more or less formal ways, by either scheduling an actual
relief (e.g., for about 2 to 10 minutes per 1/2 to 1 hour) or by designing the operator's job to include
non-viewing activities such as found in the role of communicator or police desk officer. In the 38-

3
Baker, C.A., D.F. Morris, and W.C. Steedman, "Target Recognition on Complex Displays," Human Factors, 2, No. 2
(May 1960), p. 51-61.
4
Thackray, R.I., J.P. Bailey, and R.M. Touchstone, "The Effect of Increased Monitoring Load on Vigilance
Performance Using a Simulated Radar Display," Ergonomics, 22, No. 5 (1979), p. 529-539.
5
Tickner, A.H., and E.C. Poulton, "Remote Monitoring of Motorways Using Closed-Circuit Television," Ergonomics,
n, (1968), p. 455-466.
6
Tickner, A.H., and E.C. Poulton, "Watching for People and Actions, “Ergonomics, 1_8, No. 1 (1975), p. 35-51.
7
Tickner, A.H., et al., "Monitoring 16 Television Screens Showing Little Movement," Ergonomics, lj>, No. 3 (1972),
p. 279-291.
8
Tickner, A.H., and E.C. Poulton, "Monitoring Up to 16 Synthetic Television Pictures Showing a Great Deal of
Movement," Ergonomics, 16, No. 4 (1973), p. 381-401.
monitor system referred to in the previous section, for instance, from one to three operators should be
present during the operating period of 19 hours per day. A 30-minute lunch break is required and
approximately 5-10 minutes per hour or so are provided for a rest break.9

The research on optimum viewing time indicates that vigilance can be maintained for about one hour,
and then begins to degrade noticeably. This one hour criterion is borne out by objective measures of
missed incidents and by subjective preferences reported by viewers. For example, the previously cited
research study involving eight monitor screens and a great deal of movement showed a 71% likelihood
of detecting incidents during the first hour, dropping to 62% in the second hour. The authors concluded
that the maximum desirable length of watch is about one hour. Two hours is certainly too long. Half an
hour gives us advantage over one hour. In the research study involving 16 monitor screens showing little
movement, no objective time comparisons were made, but interviews with the viewers revealed that 50
out of 65 considered that 1 hour was not too long a viewing period; 7 considered 1 hour to be about the
maximum, while 8 felt it to be too long. Opinions about the length of a regular viewing period included
durations from 1/2 to 3 hours, and 55 observers considered that it should not be more than 1 hour.10,11

In summary, the maximum recommended viewing time is about 1 hour and the maximum number of
cameras to be viewed is 16. When longer shifts are used, it is recommended that viewer personnel be
provided with 5- to 10-minute rest periods each hour or rotate jobs with other nonviewer personnel
each hour.12 The number of cameras one person can monitor is dependent upon the activity. For
instance, in an overnight environment when the activity should be very little, and the video system is set
up to push a camera image and alert the person monitoring that there is activity, the number of
cameras that can be monitored effectively is considerably more and possibly many times the number
during the day. This type of monitoring also allows for other tasks to be performed as the person would
not need to even view the monitors until directed by the system to do so.

9
Prell, J.A., "Basic Considerations for Assembling a Closed-Circuit Television System," Nuclear Regulatory
Commission, Division of Siting, Health and Safeguards Standards, Washington, DC, Technical Report No. NUREG-
0178 (May 1977). Available: NTIS, PB 268 480.
10
Prell, J.A., "Basic Considerations for Assembling a Closed-Circuit Television System," Nuclear Regulatory
Commission, Division of Siting, Health and Safeguards Standards, Washington, DC, Technical Report No. NUREG-
0178 (May 1977). Available: NTIS, PB 268 480.
11
Tickner, A.H., and E.C. Poulton, "Monitoring Up to 16 Synthetic Television Pictures Showing a Great Deal of
Movement," Ergonomics, 16, No. 4 (1973), p. 381-401.
12
file:///C:/Users/TSutton/Downloads/dot_29910_DS1.pdf