Quality in artificial lighting, good lighting for safety on roads

5 th International Symposium Topical Problems in the Field of Electrical and Power Engineering, Doctoral School of Energy and Geotechnology Kuressaare, Estonia, January 14 19, 2008 Quality in artificial lighting, good lighting for safety on roads Jelena Armas, Juhan Laugis Tallinn University of Technology arrrmas@hot.ee Abstract Street lighting and other forms of exterior lighting normally found in cities ensure that the basic requirements of residents and visitors are met: an adequate level of lighting facilitates orientation and provides security after dark. We rely on our eyes for more than 80% of the sensory impressions we register. So poor visual conditions obviously reduce the amount of information that reaches our brain. That, in road traffic, is extremely dangerous. Road lighting thus makes for greater safety at night, because it helps or even actually enables us to fill the gaps in the information we receive. Keywords Luminance, illuminance, glare, uniformity, visual acuity, luminance adaptation, overall uniformity, longitudinal uniformity, glare, threshold increment, luminance calculation, lighting measurement 1 Introduction Light is life. The relationship between light and life cannot be stated more simply than that. People nowadays spend most of the day indoors in illuminances between 50 and 500 lux. Light sets the rhythm of our biological clock but it needs to be relatively intense to have an effect on the circadian system (> 1000 lux), so for most of the time we live in chronobiological darkness. The consequences are troubled sleep, lack of energy, irritability, and even severe depression. Good lighting is important for seeing the world around us. What we want to see needs to be illuminated. Good lighting also affects the way we feel, however, and thus helps shape our quality of life. Light measuring technology has developed into an important branch of light measuring engineering. This technique provides effective solutions to a variety task in the field of light-measuring technology. Theoretical methods are usually developed for ideal conditions to determine illuminance and luminance, however in the real world other design limitations should be considered. 2 Some term of lighting technology Luminous flux Φ is the rate at which light is emitted by a lamp. It is measured in lumens (lm). Ratings are found in lamp manufacturers lists. Luminous intensity I is the amount of luminous flux radiating in a particular direction. It is measured in candelas (cd). The way the luminous intensity of reflector lamps and luminaries is distributed is indicated by curves on a graph. These are known as intensity distribution curves (IDCs). Luminous efficacy ŋ is the luminous flux of a lamp in relation to its power consumption. Luminous efficacy is expressed in lumens per watt (lm/w). Glare is annoying. It can be caused directly by luminaries or indirectly by reflective surfaces. Glare depends on the luminance and size of the light source, its position in relation to the observer and the brightness of the surroundings and background. Glare should be minimized by taking care over luminary s arrangement and shielding, and taking account of reflectance when choosing colors and surface structures for walls, ceiling and floor. Glare cannot be avoided altogether. It is especially important to avoid direct glare in street lighting as this affects road safety. Reflectance ρ indicates the percentage of luminous flux reflected by a surface. It is an important factor for calculating interior lighting. Dark surfaces call for high illuminance; lighter surfaces require a lower illuminance level to create the same impression of brightness. In street lighting, the three-dimensional distribution of the reflected light caused by directional reflectance (e.g. of a worn road surface) is an important planning factor. Luminance L indicates the brightness of an illuminated or luminous surface as perceived by the 32

human eye. It is measured in units of luminous intensity per unit area (cd/m²). For lamps, the handier unit of measurement cd/cm² is used. Luminance describes the physiological effect of light on the eye; in exterior lighting it is an important value for planning. Illuminance E is measured in lux (lx) on horizontal and vertical planes. Illuminance indicates the amount of luminous flux from a light source falling on a given surface. not possible, a maintenance factor of 0.67 is recommended for interiors subject to normal ageing and soiling; this may drop as low as 0.5 for rooms subject to special soiling. For sports facility lighting, DIN EN 12193 stipulates a maintenance factor of 0.8. Maintained value and maintenance factor define the value required on installation: maintained value = value on installation x maintenance factor. 3 Quality in lighting Depending on the use and appearance, these quality of lighting can be given different weightings. The emphasis may be on: Fig.1 Types of illuminance Eh-horizontal illuminance. This is determined by luminous flux falling on the flat horizontal surface. Ev-vertical illuminance. This is determined by luminous flux falling on the flat vertical surface. Ez-cylindrical illuminance. This is determined by luminous flux falling on the entire curved surface of an upright cylinder. Esc-semi-cylindrical illuminance. This is determined by luminous flux falling on the curved surface of an upright cylinder. Ehs-hemispherical illuminance. This is determined by luminous flux falling on the curved surface of a hemisphere standing on the surface being assessed. Maintained illuminance Em and luminance Lm depend on the visual task to be performed. Illuminance values for interior lighting are set out in the harmonized European standard DIN EN 12464-1. Illuminance and luminance values for street lighting are stipulated in DIN EN 13201-2. Sports facility lighting is covered by another harmonized European standard, DIN EN 12193. Maintained values are the values below which average values on a specified surface are not allowed to fall. Uniformity of illuminance or luminance is another quality feature. It is expressed as the ratio of minimum to mean illuminance (g1 = Emin / E) or, in street lighting, as the ratio of minimum to mean luminance (U0 = Lmin / L ). In some applications, the ratio of minimum to maximum illuminance g2. With increasing length of service, illuminance decreases as a result of ageing and soiling of lamps, luminaires and room surfaces. Under the harmonized European standards, designer and operator need to agree and record maintenance factors defining the illuminance and luminance required on installation to ensure the values, which need to be maintained. Where this is visual performance, which is affected by lighting level and glare limitation, visual comfort, which is affected by color rendering and harmonious brightness distribution, visual ambience, which is affected by light color, direction of light and modeling Fig.2 Weightings of lighting 3.1 Lighting level maintained illuminance and luminance Lighting level is influenced by illuminance and the reflective properties of the surfaces illuminated. It is a defining factor of visual performance. Some examples of reflectance: white walls up to 85% light-colored wood paneling up to 50% red bricks up to 25%. The lower the reflectance and the more difficult the visual task, the higher the illuminance needs to be. Maintained illuminance is the value below which the average illuminance on the assessment plane is not allowed to fall. With increasing length of service, illuminance is reduced owing to ageing and soiling of lamps, luminaires and room surfaces. To compensate for this, a new system needs to be designed for higher illuminance (value on installation). The reduction is taken into consideration by a maintenance factor: maintained illuminance = maintenance factor x illuminance on instalation. 33

The maintenance factor depends on the maintenance characteristics of lamps and luminaries, the degree of exposure to dust and soiling in the room or surroundings as well as on the maintenance programmed and maintenance schedule. In most cases, not enough is known at the lighting planning stage about the factors that will later impact on illuminance, so where a maintenance interval of three years is defined, the maintenance factor required is 0.67 for clean rooms and as low as 0.5 for rooms subject to special soiling (e.g. smoking rooms). The surface on which the illuminance is realized is normally taken as the evaluation plane. Recommended heights: 0.75 m above floor level for office workplaces, max. 0.1 min circulation areas. The maintained illuminances required for indoor workplaces are defined in DIN EN 12464-1 for different types of interior, task or activity. Examples: circulation areas 100 lx office 500 lx For sports lighting, reference planes (at floor/ground level) and illuminance requirements are set out for different types of sport in the harmonized European standard DIN EN 12193. Illuminance is the variable used for planning interior lighting because it is easy to measure and fairly straightforward to compute. Luminance Determining luminance L (cd/m²) entails more complex planning and measurement. For street lighting, luminance is an essential criterion for assessing the quality of a lighting system. What motorists see is the light reflected in their direction from the perceived road surface (the material dependent and directional luminance). Since the reflectance of road surfaces is standardized and a single observation point has been defined as standard, luminance is the variable normally used for planning street lighting. The illumination of a street depends on the luminous flux of the lamps, the intensity distribution of the luminaries, the geometry of the lighting system and the reflectance of the road surface. The quality features of street lighting are listed in DIN EN 13201-2. Recommended values: local service street 7.5 lx main thoroughfare 1.5 cd/m² car park 15 lx. 4. Glare Direct glare is caused by excessive luminance e.g. from unsuitable or inappropriately positioned luminaries or from unshielded general-diffuse lamps. Glare causes discomfort (psychological glare) and can also lead to a marked reduction in visual performance (physiological glare); it should therefore be limited. 4.1. The TI method in street lighting Every motorist is aware of the dangers of glare in street lighting and its implications for road safety. Effective limitation of physiological glare is therefore an important requirement for good street lighting. The method used to limit glare in street lighting is based on the physiological effect of glare and demonstrates the extent to which glare reduces the eye s threshold of perception. In outdoor lighting, physiological glare is assessed by the TI (Threshold Increment) method. The TI value shows in percent how much the visual threshold is raised as a result of glare. The visual threshold is the difference in luminance required for an object to be just perceptible against its background. Example: Where street lighting is glare free, the eye adapts to the average luminance of the road L. A visual object on the roadway is just perceptible where its luminance contrast in relation to its surroundings is L0 (threshold value). Where dazzling light sources occur in the visual field, however, diffuse light enters the eye and covers the retina like a veil. Although the average luminance of the road remains unchanged, this additional veiling luminance Ls causes the eye to adapt to a higher level L + Ls. An object with a luminance contrast of L0 in relation to its surroundings is then no longer visible. Where glare occurs, luminance contrast needs to be raised to LBL for an object to be perceptible. On a road of known average roadway luminance L, the increment LBL - L0 can be used as a yardstick for the impact of glare. The percentage rise in threshold values TI (Threshold Increment) from L0 to LBL has been adopted as a measure of physiological glare and is calculated on the basis of the following formula: TI/%= (( LBL - L0)/ L0)*100 Fig.3 Longitudinal uniformity 4.2.The UGR method in indoor lighting In indoor lighting, psychological glare is rated by the standardized UGR (Unified Glare Rating) method. This is based on a formula, which takes account of all the luminaries in a lighting system, which contributes to a sensation of glare. Glare is assessed using UGR tables, which are based on the UGR formula and are available from luminary s manufacturers. UGR=8 log 10 0.25/LbΣLi²Ωi/pi² Lb background or also surrounding luminance in cd/m² 34

L luminance of the glaring parts of a luminaire in the direction of the observer s eye in cd/m² Ω solid angle of the glaring parts of a luminaire p position factor of an individual luminaire i number of the extracted light source 5. Measuring lighting systems In lighting engineering, measurements are taken to check lighting proposals, check the condition of existing lighting systems to determine whether maintenance or refurbishment are required, compare different lighting systems. Standards and regulations set out stipulations to ensure that measurement and evaluation methods are standardized. Important variables are: illuminance E, e.g. as horizontal illuminance Eh, as vertical lluminance Ev, as cylindrical illuminance Ez or semi-cylindrical illuminance Ehz. luminance L, e.g. in street lighting, tunnel lighting or interior lighting, reflectance ρ, e.g. of ceilings, walls, floors, in workplace interiors and sports halls, the reflective properties of road surfaces, e.g. in street and tunnel lighting, line voltage U and/or ambient temperature ta for lighting systems with lamps whose luminous flux is dependent on the service voltage and/or the room or ambient temperature. Fig. 4 Horizontal illuminances on the working plane In practice, the variable measured most frequently is illuminance. For this, instruments with a relative spectral sensitivity comparable to that of the human eye V(λ) are used. Oblique incident light needs to be measured in line with the cosine law. When preparing photometric procedures, the following need to be established: geometric dimensions of the lighting system, type of system/nature of interior and activity, variables to be measured and location of measuring points, general condition of the system, e.g. age, date of last cleaning and last lamp replacement, degree of soiling. Before measurements are taken, lamps should be left on long enough for the system to reach a steady state and interference by extraneous light (e.g. daylight influencing interior or vehicle lighting, shop window or advertising lighting influencing outdoor lighting) should be eliminated. Fig.5 Luminance L of the road surface/roadway Interference due to obstacles or shadows cast by persons taking measurements must also be avoided. For illuminance measurements, the ground or floor area of the installation in question should be divided into preferably square patches of equal size. To avoid obtaining only maximum values, e.g. directly under luminaires, the measurement grid thus formed should not reflect the modular dimensions of the luminaries arrangement. However, symmetrical features of lighting system, room or outdoor space can be usefully employed to reduce the number of measurements required. Measurements are presented in tables. A graphic representation of illuminances in isolux curves is obtained by joining up points of equal illuminance. To determine mean illuminance E, the individual measurements are added together and divided by the number of points at which measurements are taken. The uniformity of illuminance g1 is the quotient of the lowest illuminance value ascertained Emin and the mean illuminance E calculated. Uniformity g2 is the ratio of Emin to the highest illuminance value ascertained Emax. A record of each measurement should be kept, documenting, for example, not just the values themselves but also the ambient conditions, details of lamps, luminaries and the geometry of the lighting system. 6 Calculating street luminance according to EN 13201-3:2003 The street luminance we can calculate with program LMK 2000 Image capturing and processing. In the following paragraph, the evaluation of the image data in the LMK2000 software is described. A precondition for the evaluation is the transfer of the raw data (CR2 files) from the camera into a directory on the PC with LMK2000 evaluation software and with the 35

standard Canon software, which will start automatically when the camera is connected. As soon as the Canon software is started, the image files can be chosen in the camera and then copied into a target file on the PC. After having chosen the HighDyn measurement either via the menu item Capture HighDyn or the button, the dialog shown opposite will appear. By pressing the button the path (the target directory of the CR2 files) can be selected. All CR2 files contained in the folder chosen will be displayed. Now, an image file can be selected, and the directory is opened either via Open or by a double click on the file. The images will be listed (listing may take a few seconds during the first display). The images belonging to one exposure series will be displayed in one line. In the lower right-hand section of the bitmap, the internal image data (header information) will appear. If the serial number of the camera does not correspond with the serial number of the calibration data, the serial number will be backlit in red. Choosing the option Single Image, the single files will be listed without being classified according to their serial numbers. Then, via the multiselection option (using the ctrl or the shift key for marking), those files can be selected which shall be combined to one capture. In doing so, attention has to be paid to the fact that all captures must be taken using one and the same aperture value and that the position of the camera is not changed. The image data will then be evaluated only by means of the integration times. Then By pressing the OK key, the selected images will be taken over. By pressing the Cancel key, no image will be taken over. (Here, a Preview function is planned to be installed in the future.), the selected image files will be read in and combined to form a luminance image or also a color picture. During HighDyn capture, the data of the single images will be displayed (image number, integration time, aperture and file name). If the message Last camera image overdriven is displayed after reading in the last capture, this means that there are still some image regions, which are overdriven in all captures, which makes it impossible for them to be evaluated. In this case, the user will have to apply shorter integration times or larger aperture numbers or, if necessary, gray filters so as to allow evaluation. Fig.8 Image reading A summary of all settings can be found under the menu item File Details of Measurement. Fig.9 Details of measurment Via the menu item Capture HighDyn directory, a complete directory of Canon-RAW (CR2 files) can be converted into luminance images. To do this, proceed as follows: Fig.6 Image files Fig.7 Option Single Image Select menu item Capture HighDyn directory. Then, select an image directory according to the normal evaluation Capture HighDyn, and select the first group or file. Afterwards, all file groups are read in and converted into one luminance image each. The images will be saved in the folder CanonResult under the name of a file (e.g. IMG_6279) within the installation file as PF image file. By means of the measuring series functions, these images can then be read in and evaluated. By pressing the OK key, the conversion will be terminated at once. The entire directory will be converted without interaction. 36

Fig.10 File selecting After the image data have been read in and converted into a luminance image, all evaluation functions offered by the LMK2000 software will be available for the further processing of the data. For calculating the luminances, the information of the R, G, B pixels (red, green, blue pixels) will be read out of the data of the Bayer structure of the CMOS color sensor of the camera. 4 pixels each time (R, G1, B and G2) form a macropixel for which a luminance value will be calculated. The luminance image has a size of 1728(H) x 1152(V) pixels. The adaptation of the relative spectral sensitivity of the system to the V(λ)-function will not be effected using additional filters, but through numerical matrixing. The assignment of the image data to specific luminance values will not be carried out at standard illuminant A, any deviating spectral distributions can possibly result in more serious measuring errors. Using the menu item Image processing. Isolines, a new image can be calculated. In this image, regions of different luminances are either closed off from each other by lines, or those regions have been marked differently. When this menu item is called, a dialogue will be opened which permits the source and the target images of the calculation to be fixed. If necessary, a new target image can be created. The size of the target image will automatically be adapted to the size of the source image. Then, a dialogue will be opened where the parameters of the calculation can be set. On the top left in the dialogue, the Minimum and the Maximum of the luminance of the source image will be indicated. These two parameters are useful for the indication of the luminance thresholds. On the right-hand side of the dialogue, the Number of thresholds can be indicated. The list below contains the values of these thresholds. Using the buttons to the left of the list, the user can influence the thresholds: These values can be indicated either in absolute (luminance) values or in relative values in percent. In the latter case, the user may choose whether the percentage shall refer to the range between minimum and maximum (Image minimum. Note) or between zero and maximum (Image minimum. Ignore). If the option Thresholds.Default setting is chosen, the thresholds will be evenly set automatically. If the option Thresholds. Edit is selected, the numerical values in the list can be changed manually. The graphical representation can be made either in the form of areas of the same brightness for the luminance s within an interval between two luminance thresholds (Representation. Areas) or in the form of demarcation lines between regions of different luminance s at the point where the luminance thresholds are exceeded (Representation. Lines). In many case, the demarcation lines between regions of different brightness are greatly frayed. Using Smooth image, these frayed contours can be smoothed slightly. The higher the value entered is, the greater the smoothing effect will be. In the default setting, the isolines will be drawn in the resulting image in the form of a line one pixel wide. If this image is represented at an enlargement of 66% or less, the lines calculated will then appear as broken lines. These discontinuities can be removed using Draw bold lines. Then, the lines inserted in the resulting image will get a bigger width. If you want to label the areas or lines with the thresholds chosen, the option Labelling along the line(s) marked can be used. For this, lines must be inserted in the target image and then marked. By pressing the button Execute, the calculation of the target image is started. At the same time, the desired labelling is performed. If Representation: Lines has been chosen, the corresponding thresholds will be displayed in the intersection points of the lines drawn in and the demarcation lines calculated. If Representation: Areas has been chosen, then the corresponding values will be displayed in the middle of the line section, by an area of the same brightness. Having terminated the calculation of the isorepresentation, the auxiliary lines drawn in can be removed again. The labelling, however, will remain. If the option All lines of same brightness is switched off, also the points of the demarcation lines between regions of different luminance s will have different values. Using the option Invert representation, the user can decide whether black lines on white ground or white lines on black ground shall be displayed in the line representation. The same procedure can be applied to the representation of areas. The measuring protocols of the LMK2000 device serve to store the images that have been captured, as well as measuring results in order to make them available at any later time. The data will be stored in the HDF5-file format. Protocols of the LMK96/98 device saved in the PRT-format can be read but not written. 37

The following data can be stored in the HDF5- protocol files: Images (camera, luminance and all additional evaluation images) Settings relating to the image display (color palettes used, scales, enlargements) Measuring regions Measuring values (luminance statistics, graphics, inspector settings relating to luminance objects, etc.) Additional data such as capture parameters, user s name, etc. Comment on the measurement as RTF-text By using the menu item File.Options, you can open a dialogue where the protocol settings for loading, storing and creating new protocols can be realized individually. For each of the actions New, Load and Save the dialogue contains a separate register card. The possible settings are explained in the paragraphs concerning the loading and storing of protocols and the new creation of protocols. 10 Conclusion The LMK luminance measuring cameras capture images and calculate from the image data (gray values per pixel) the luminance s in the respective points of the road. Thus, these luminance images contain information about the light and the geometry, thus permitting any measuring tasks to be solved in a very simple and time-efficient way. For the photometric measurement of long stretches of roads, image series can be captured. When the lighting measurement principles are properly applied to increase visibility on the roadways, they can provide social and economic benefits to the public, including: reduction of nighttimes accidents; aid to police protection; facilitation of traffic flow; promotion of businesses; inspiration of community growth; safety for pedestrians. All traffic and lighting engineers should be encouraged to support research and development in this area as it will benefit everyone for years to come. References [1] Lighting Manual, Philips Lighting B.V., 1993, pp. 273-347 [2] J.E. Kaufman, J.F. Christensen, IES Lighting Handbook, 1987, pp. 14-1 - 14-35 [3] Good Lighting for Safety on Roads, Paths and Squares, Fördergemeinschaft Gutes Licht, 2000, pp 3-11 [4] DIN 5044 Stationary traffic lighting Street lighting for automobile traffic Part 1: General requirements and recommendations Part 2: Calculation and measurement [5] EN 13201 Street lighting draft Part 1: Quality criteria with appendix for lighting class selection Part 2: Calculation of quality criteria Part 3: Methods for measuring quality criteria [6] Illuminance measurements of Roadways, Hans Peter Grieneisen, Aline S.P.Timmins, André S. Sardinha, Iakyra B. Couceiro, 2006 [7] Roadway lighting design methodology and evalution, Olkan Cuvalci, Burga Ertas, Martch 2000 [8] Luminance analysers-what they are and how do they work, Wolf, Stefan; Gall, Dietrich, 2002, pp. 1-4 [9] Conception/design of lighting systems for urban traffic roads, Cornel Bianchi, Camelia Burlacu, 2001 [10] Applied Image Resolved Light- and Color Measurment Introduction and Application Examples, U. Krüger, F. Schmidt, TechnoTeam Bildverarbeitung GmbH, Germany, 2001 [11] Image-resolving luminance measurement using luminance measuring cameras type LMK96/98 (Video analyzers), TechnoTeam Bildverarbeitung GmbH, Germany, 2001, pp. 3-4, pp. 8-18 [12] CCD Technology Primer, www.dalsa.com, pp 17-23 [13] Solid State image. Terminology, Microelectronics technology division Rochester, New York 14650, pp. 4-30 [14] Applications of the image resolved light and color measurement, Schmidt, Franz; Krüger, Udo TechnoTeam Bildverarbeitung GmbH, 2001 [15] A Light Hierarchy for Fast Rendering of Scenes with Many Lights, Eric Paquette, Pierre Poulin, and George Drettakis, 1999 [16] Environmental effects of roadway lighting, Carl Shaflik, 1999 [17] European Road Lighting Technologies, US DOT FHWA International Technology Exchange Program Office of International Programs. pp 9-18 [17] LMK 2000 Operation Manual, Techno Team Bildverarbeitung GmbH. 2003 38