A simple safety framework for laser eye dazzle, based on a complex model developed from human subject experiments, is proposed to address the urgent need for guidance within international laser safety standards. Maximum Dazzle Exposure (MDE) safety limits are derived that set the laser irradiance at the eye above which an object cannot be visually detected. A newly defined concept of dazzle level accounts for the extent of visual obscuration, and different ambient light levels are accommodated by determining safety limits for night, dusk/dawn, and day conditions. The resulting table of MDE values allows dazzle effects to be quantified in simple safety calculations across a wide range of scenarios. This safety framework is intended to empower the laser safety community to understand and quantify the impacts of laser eye dazzle, specify protection measures for those at risk, and assure the safety and effectiveness of laser dazzle devices.

Laser eye dazzle—the temporary impairment of human performance caused by visible wavelength laser light—is increasingly being encountered in civilian and military domains. Commercial aircraft are being maliciously targeted by high power handheld lasers that are capable of causing visual disruption to pilots at ranges of many kilometers, with almost 20 incidents per day reported during 2017 in the US alone.1 Security forces are also increasingly deploying laser dazzle as a nonlethal option to warn and determine intent,2 providing an intermediate step between “shouting” and “shooting.”

There is an urgent need for a safety framework that allows the impacts of laser eye dazzle to be understood and quantified. Cases of permanent eye damage caused by lasers are still relatively rare,3,4 with the risks of eye damage from laser systems being well understood and comprehensively documented in international laser safety standards.5,6 However, despite the significantly greater frequency of laser eye dazzle events, there is an absence of equivalent guidance for laser eye dazzle effects in these standards.

With the aim of enhancing international laser safety standards, the authors established their own methodology for assessing laser eye dazzle effects in 2015 (Ref. 7) and are now proposing a major revision. This earlier work recommended additional human subject experiments to improve accuracy, and a refined calculation approach to improve simplicity. Following the completion of the required human subject experiments,8 the authors are now able to recommend an approach that is indeed improved in accuracy and simplicity.9 

This paper presents the scientific background to the new laser eye dazzle safety framework, together with the full text of the proposed framework as an annex.10 This framework is designed to be a self-contained summary of what laser eye dazzle is, what effects it has on human performance, what the main contributors are to its severity, how to mitigate it, and how to predict its effects with simple calculations. The main body of this paper serves as a technical reference for the safety framework, detailing the derivation of the proposed Maximum Dazzle Exposure (MDE) safety limits with respect to the experimental data and underlying computer model.

Integral to the simple calculation of laser eye dazzle effects is the concept of MDE. Laser eye dazzle causes part of the visual scene to be obscured by the appearance of a “dazzle field”—a bright, saturated region of vision as illustrated in Fig. 1. The MDE is the laser irradiance at the eye above which an object cannot be visually detected through this dazzle field. At laser irradiances higher than the MDE, the dazzle field prevents the observer from detecting the object, while at irradiances lower than the MDE the observer is able to detect the object. MDE is analogous to the established concept of Maximum Permissible Exposure (MPE) which sets the laser irradiance limit above which there is a risk of permanent eye damage.

FIG. 1.

Simulated image of the dazzle field caused by laser eye dazzle.

FIG. 1.

Simulated image of the dazzle field caused by laser eye dazzle.

Close modal

It is not possible to specify a single MDE to account for all scenarios. Instead, a table of MDE values has been designed to cover the major contributors to the severity of laser eye dazzle, as will now be detailed.

It is proposed that MDE values are specified for a set of dazzle levels (DLs). These dazzle levels quantify the extent of visual obscuration experienced, ranging from Very Low (2° full angle dazzle field, centered on the laser source) where a laser beam can be seen but it only obscures a minor extent of the visual field, to Low (10°), Medium (20°), and High (40°) which represent increasing extents of visual disruption (see Fig. 2). The MDE will be higher for higher DLs, and selection of the appropriate MDE will be dictated by the DL that is deemed acceptable according to the visual requirements of the observer for the task being completed.

FIG. 2.

Illustration of the visual extent of the four dazzle levels (image scene horizontal field of view is 40°).

FIG. 2.

Illustration of the visual extent of the four dazzle levels (image scene horizontal field of view is 40°).

Close modal

It is proposed that MDE values are specified for ambient luminance levels approximating night (0.1 cd m−2), dusk/dawn (10 cd m−2), and day (1000 cd m−2) conditions, with laser eye dazzle more readily achieved as the ambient light level decreases from day to night. The MDE will be greater for higher ambient light levels, and selection of the appropriate MDE is therefore dictated by the ambient luminance conditions under which laser exposure is anticipated.

It is proposed that MDE values should be scaled depending upon the laser wavelength being considered, according to the eye’s photopic sensitivity,11,12V(λ). The eye is less sensitive to blue (e.g., V(450 nm) = 0.0647) and red (e.g., V(650 nm) = 0.1193) wavelengths compared to green (e.g., V(550 nm) = 0.9890), meaning that MDE values will be higher for wavelengths of lower sensitivity as more irradiance is required to achieve the same effect. All MDE values should therefore be divided by V(λ) for the laser wavelength to give the appropriate MDE.

Table I illustrates the structure of the proposed MDE table to account for the primary contributors to the severity of laser eye dazzle. It provides a range of values which depend upon the acceptable dazzle level and the ambient light level, with all values being divided by V(λ) to account for the laser wavelength dependence.

TABLE I.

Structure of the proposed MDE table.

Dazzle level MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  —  —  —  ÷ V(λ
Low  —  —  — 
Medium  —  —  — 
High  —  —  — 
Dazzle level MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  —  —  —  ÷ V(λ
Low  —  —  — 
Medium  —  —  — 
High  —  —  — 

As the MDE represents the laser irradiance that prevents visual detection of an object, the object being viewed must be defined. A standard object size of 0.08° with 60% contrast has been chosen, based upon the authors’ human subject experiments8 which used a tumbling-E letter orientation task with 60% contrast characters of 0.4° size (0.08° bar size) viewed monocularly in negative contrast (i.e., dark letters on a light background). Detecting a 0.08° object is approximately equivalent to detecting a 5 m long car at a range of 3.5 km or a 0.5 m human torso at 350 m.

Table II populates the MDE table with values from the human subject experiments, as well as from the updated computer model. Experimental data from the authors’ earlier work8,13 were averaged across all laser wavelengths (up to eight) for each relevant combination of ambient luminance and object size, with the photopic luminous efficiency being factored out of the calculations. The experimental irradiance ranges given are the average value plus and minus one standard deviation, with footnote “a” denoting values that were extrapolated beyond the irradiance levels used in the actual experiment. Details of the averaging and extrapolation can be found in the analysis provided with the data set for this paper.14 

TABLE II.

Human subject experiment and modeled MDE values for monocular viewing of the negative contrast 0.08° bar width, 0.4° E letter (60% contrast).

Dazzle level Data source MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  Expt.  —  0.1–0.4a  14–18  ÷ V(λ
Model  0.0001  0.4  21 
Low  Expt.  —  8–19  688–5619a 
Model  0.003  17  1039 
Medium  Expt.  —  43–106  3694–66 734a 
Model  0.01  74  4495 
High  Expt.  —  221–636  19 829–792 633a 
Model  0.04  281  16 950 
Dazzle level Data source MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  Expt.  —  0.1–0.4a  14–18  ÷ V(λ
Model  0.0001  0.4  21 
Low  Expt.  —  8–19  688–5619a 
Model  0.003  17  1039 
Medium  Expt.  —  43–106  3694–66 734a 
Model  0.01  74  4495 
High  Expt.  —  221–636  19 829–792 633a 
Model  0.04  281  16 950 
a

Values that were extrapolated beyond the irradiance levels used in the actual experiment.

Model data are based upon the previously reported calculation method7,8 with inputs as given in Table III. This method uses established models for human eye scatter and visual detection contrast thresholds and includes a laser exposure calibration factor derived from the human subject laser experiments. An age of 40 was chosen for these calculations, together with a dark eye pigmentation (=0.5), which was the approximate average profile of subjects used in the experiments. Again, these calculations can be found in detail in the supporting data set.14 

TABLE III.

Input parameters to MDE model calculations for visual detection of a 0.08° object.

Parameter Value
Observer age (years)  40 
Eye pigmentation  0.5 
Object size for detection (deg)  0.08 
Object contrast  0.60 
Photopic luminous efficiency 
  Night  Dusk/dawn  Day 
Ambient luminance (cd m−2 0.1  10  1000 
Contrast thresh., monocular negative  0.5794  0.0955  0.0730 
Contrast thresh., binocular negative  0.3408  0.0562  0.0430 
Contrast thresh., binocular positive  0.4908  0.0690  0.0433 
Parameter Value
Observer age (years)  40 
Eye pigmentation  0.5 
Object size for detection (deg)  0.08 
Object contrast  0.60 
Photopic luminous efficiency 
  Night  Dusk/dawn  Day 
Ambient luminance (cd m−2 0.1  10  1000 
Contrast thresh., monocular negative  0.5794  0.0955  0.0730 
Contrast thresh., binocular negative  0.3408  0.0562  0.0430 
Contrast thresh., binocular positive  0.4908  0.0690  0.0433 

It can be seen that the model provides values that are broadly within the range of the experimental data. As the experimental data become extrapolated, primarily at day ambient luminance levels, the model continues to be near the experimental data range although towards the lower end. As 2546 μW cm−2 represents the MPE for a 0.25 s visible wavelength exposure, it is indeed preferable that the model errs on the side of caution for these high irradiance levels by setting a safety limit towards the lower end. The day MDE values in Table II that exceed the MPE were extrapolated beyond the irradiance values used in the actual experiment. The experiment never exceeded 1000 μW cm−2, which is the MPE for a 10 s visible wavelength exposure.

The human subject experiments did not evaluate the 0.4° E at night ambient luminance levels because that target was too near threshold and could not be reliably discriminated by many of the subjects. Therefore, Table IV presents MDE values for the 0.16° bar size, 0.8° E character (70% contrast) that was evaluated at these ambient levels. As well as supporting the accuracy of the model’s predictions at dusk/dawn and day levels, these data show that the model is also capable of predicting exposure levels at night ambient luminance levels with a good degree of accuracy.

TABLE IV.

Human subject experiment and modeled MDE values for monocular viewing of the negative contrast 0.16° bar width, 0.8° E letter (70% contrast).

Dazzle level Data source MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  Expt.  0.002–0.004a  0.4–1.5a  38–78  ÷ V(λ
Model  0.005  1.1  55 
Low  Expt.  0.12–0.16a  20–46  1562–25 588a 
Model  0.24  56  2698 
Medium  Expt.  0.69–0.75  103–217  7723–309 383a 
Model  1.0  241  11 668 
High  Expt.  3.6–4.1  525–1054a  38 200–3 740 664a 
Model  3.9  907  43 998 
Dazzle level Data source MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  Expt.  0.002–0.004a  0.4–1.5a  38–78  ÷ V(λ
Model  0.005  1.1  55 
Low  Expt.  0.12–0.16a  20–46  1562–25 588a 
Model  0.24  56  2698 
Medium  Expt.  0.69–0.75  103–217  7723–309 383a 
Model  1.0  241  11 668 
High  Expt.  3.6–4.1  525–1054a  38 200–3 740 664a 
Model  3.9  907  43 998 
a

Values that were extrapolated beyond the irradiance levels used in the actual experiment.

Having demonstrated the ability of the model to predict MDE values accurately, the model was next applied to the more representative scenario of an object being viewed binocularly. This involved changing the contrast threshold values for visual detection as shown in Table III, and reverting back to the 0.08° object with 60% contrast. After calculating MDE values for binocular negative contrast and binocular positive contrast object detection, the two values for each entry were averaged to give the MDE table shown in Table V.

TABLE V.

Modeled MDE values for binocular detection of a 0.08° object with 60% contrast.

Dazzle level MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  0.00074  0.58  38  ÷ V(λ
Low  0.036  28  1859 
Medium  0.16  122  8042 
High  0.59  462  30 326 
Dazzle level MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  0.00074  0.58  38  ÷ V(λ
Low  0.036  28  1859 
Medium  0.16  122  8042 
High  0.59  462  30 326 

Following the derivation of these values, some rounding was conducted to simplify presentation, giving the values in Table VI which are offered as the proposed table of MDE values to be used in laser safety standards.

TABLE VI.

Proposed standard table of MDE values.

Dazzle level MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  0.001  0.6  40  ÷ V(λ
Low  0.04  30  2000a 
Medium  0.16  120  8000a 
High  0.6  450  30 000a 
Dazzle level MDE (μW cm−2) at
Night Dusk/dawn Day
Very low  0.001  0.6  40  ÷ V(λ
Low  0.04  30  2000a 
Medium  0.16  120  8000a 
High  0.6  450  30 000a 
a

Values that exceed the visible wavelength MPE of 1000 μW cm−2 for a 10 s exposure. The MPE could also be exceeded in other cases when dividing by small V(λ) values.

Recognizing that these MDE values will be higher for all laser wavelengths outside the eye’s peak response at 555 nm (where V(λ) = 1), they indicate that at night, only around 1 nW cm−2 is required for a Very Low DL, while a High DL can be achieved with around 0.6 μW cm−2. That same irradiance would only induce a Very Low DL at dusk/dawn, where around 450 μW cm−2 would be required for a High DL. During the day, a Very Low DL can be achieved with around 40 μW cm−2, but greater levels of dazzle require irradiances that exceed the visible wavelength MPE—i.e., the observer would be at risk of eye damage if exposed to these levels of dazzle.

The final MDE table should be taken only as an approximate guide to the laser irradiances required to cause the stated dazzle levels at the given ambient light levels. MDE values would be expected to be higher for detection of larger or higher contrast objects, and lower for the detection of smaller or lower contrast objects than the 0.08° 60% contrast object assumed here. MDE values would also be expected to be higher for younger observers and lower for older observers, compared with the 40 year old observer assumed here. It should also be noted that a 40 year old with dark eyes who is detecting a 0.08° 60% contrast object (i.e., the precise case presented here) would also likely experience dazzle different from that predicted by this table as there is variance among the population in terms of eye scatter and how laser dazzle impacts visual performance. However, the MDE table does give a useful approximation of these visual effects and allows a rapid understanding of the likely impact of given laser irradiances.

MDE values can also be used to predict a safe distance for operating in the presence of laser dazzle, known as the Nominal Ocular Dazzle Distance (NODD).7 The NODD is defined as the distance beyond which the irradiance delivered by a laser is below the MDE. The NODD is therefore the dazzle equivalent to the Nominal Ocular Hazard Distance (NOHD) that is used to quantify the safe distance to operate from a laser system to avoid any risk of eye damage. NODD can be derived from MDE values in exactly the same way as NOHD is calculated from MPE values, with details of the calculation and some worked examples provided in the safety framework annex.10 

The annex10 includes some visualizations of the dazzle field caused by laser eye dazzle, as also shown in Fig. 1. These visualizations are based upon an extended version of a previously reported technique.15 They use the new scatter function generated by the human subject experiments8 and incorporate additional visual effects for added realism, such as a representation of the ciliary corona.16 Just like the MDE table, they cannot give a precise replication of the experience of laser eye dazzle, but they are intended to give a rapid appreciation of what laser eye dazzle looks like and how its severity is affected by laser irradiance, ambient light level, and laser wavelength.

As with the main experiment that validates this work,8,13 the authors have made available the complete data and analysis from this paper to encourage independent verification of the methodology.14 A spreadsheet calculator has also been published17 to facilitate calculation of MDE and NODD values for specific scenarios, together with optical density requirements for laser eye protection.

The authors are grateful to their many colleagues across UK MOD and US DOD, and the wider research community who have contributed over many years to the underlying research that has made this safety framework possible. DSTL/JA101326. Content includes material subject to © Crown copyright (2018), Dstl. This material is licensed under the terms of the Open Government Licence except where otherwise stated. To view this licence, visit http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3 or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU, or email: psi@nationalarchives.gsi.gov.uk.

1.
LaserPointerSafety.com
, see http://laserpointersafety.com/news/news/other-news_files/d5fca5f3330bef52f1c5465b03502776-569.php for “US: 6,753 laser/aircraft incidents reported to FAA in 2017; 9% lower than 2016” (
2018
).
2.
Joint Non-Lethal Weapons Program
, see http://jnlwp.defense.gov/Portals/50/Documents/Press_Room/Fact_Sheets/NL_Optical_Distracters_Fact_Sheet_May_2016.pdf for “Non-lethal optical distracters fact sheet” (May
2016
).
3.
D. B.
Gosling
,
J. B.
O’Hagan
, and
F. M.
Quhill
, “
Blue laser induced retinal injury in a commercial pilot at 1300 ft
,”
Aerosp. Med. Hum. Perform.
87
,
69
70
(
2016
).
4.
J.
Marshall
,
J. B.
O’Hagan
, and
J. R.
Tyrer
, “
Eye hazards of laser ‘pointers’ in perspective
,”
Br. J. Ophthalmol.
100
,
583
584
(
2016
).
5.
American National Standards Institute
, ANSI Z136.1-2014—American National Standard for Safe Use of Lasers (
2014
).
6.
European Committee for Electrotechnical Standardization
, EN 60825-1:2014—Standard, Safety of laser products—Part 1: Equipment classification and requirements (
2014
).
7.
C. A.
Williamson
and
L. N.
McLin
, “
Nominal Ocular Dazzle Distance (NODD)
,”
Appl. Opt.
54
,
1564
1572
(
2015
).
8.
C. A.
Williamson
,
L. N.
McLin
,
J. M.
Rickman
,
M. A.
Manka
,
P. V.
Garcia
,
W. T.
Kinerk
, and
P. A.
Smith
, “
Wavelength and ambient luminance dependence of laser eye dazzle
,”
Appl. Opt.
56
,
8135
8147
(
2017
).
9.
C. A.
Williamson
and
L. N.
McLin
, “
Laser eye dazzle safety framework
,” in
Proceedings of ILSC 2017—International Laser Safety Conference
,
Atlanta, GA
, March 2017, (
Laser Institute of America
,
Orlando, FL
,
2017
), pp.
164
165
.
10.
See supplementary material at https://doi.org/10.2351/1.5029384 for complete text of the proposed laser eye dazzle safety framework.
11.
CIE (International Commission on Illumination)
, “
Fundamental chromaticity diagram with physiological axes—Part 2: Spectral luminous efficiency functions and chromaticity diagrams
,”
CIE Report CIE 170-2:2015
, (
2015
); available at http://www.cie.co.at/publications/fundamental-chromaticity-diagram-physiological-axes-part-2-spectral-luminous-efficiency.
12.
CVRL Database
, see http://www.cvrl.org/ for CIE “physiologically-relevant” luminous efficiency functions consistent with the Stockman & Sharpe cone fundamentals, 2-deg functions.
13.
C. A.
Williamson
,
L. N.
McLin
,
J. M.
Rickman
,
M. A.
Manka
,
P. V.
Garcia
,
W. T.
Kinerk
, and
P. A.
Smith
, see for dataset “Data to accompany paper: Wavelength and ambient luminance dependence of laser eye dazzle,” figshare (
2017
).
14.
C. A.
Williamson
and
L. N.
McLin
, see for dataset “Data to accompany paper: Determination of a laser eye dazzle safety framework,” figshare (
2017
).
15.
C. A.
Williamson
, “
Simple computer visualization of laser eye dazzle
,”
J. Laser Appl.
28
,
012003
(
2016
).
16.
G. S.
Spencer
,
P.
Shirley
,
K.
Zimmerman
, and
D. P.
Greenberg
, “
Physically-based glare effects for digital images
,” in
SIGGRAPH 95 Conference Proceedings, Annual Conference Series, ACM SIGGRAPH
,
Los Angeles, California
, August 1995, edited by
R.
Cook
(
ACM
,
New York
,
1995
), pp.
325
334
.
17.
C. A.
Williamson
and
L. N.
McLin
, see for dataset “Laser safety calculator for laser eye dazzle effects and protection,” figshare (2017).

Supplementary Material