{
  "id": "luminous-intensity",
  "title": "Luminous intensity",
  "lang": "en",
  "source_lang": "en",
  "summary": "Luminous intensity measures the visually weighted light emitted in a particular direction per unit solid angle, expressed in candelas.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [
    "Photometric intensity"
  ],
  "generated_at": "2026-11-06",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/luminous-intensity/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/luminous-intensity/",
      "json": "https://aiwiki.page/zh/luminous-intensity.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Luminous intensity is the amount of visible light emitted by a source in a specified direction per unit solid angle, weighted according to a standardized model of human visual sensitivity. Its unit is the candela (cd), one of the seven base units of the International System of Units (SI). A source has a luminous intensity of one candela in a direction when it emits one lumen of luminous flux per steradian about that direction.\n\nThe quantity describes directional emission, not total light output or the brightness of an illuminated surface. A narrow-beam spotlight can have a much higher peak luminous intensity than a broadly emitting lamp with the same total output. This distinction makes luminous intensity important in the specification of vehicle headlamps, signal lights, projectors, street lighting, and other systems in which the direction of light matters.\n\nLuminous intensity belongs to photometry, the measurement of light in terms of standardized visual response. Its definition connects the geometry of radiation with the wavelength-dependent sensitivity of the eye. Its practical determination therefore requires both optical measurements and a carefully maintained relationship between photometric units and physical measurements of radiant power."
    },
    {
      "heading": "Definition and physical meaning",
      "text": "The luminous intensity \\(I_{\\mathrm v}\\) of a source in a particular direction is defined by\n\n\\[I_{\\mathrm v}=\\frac{\\mathrm d\\Phi_{\\mathrm v}}{\\mathrm d\\Omega},\\]\n\nwhere \\(\\Phi_{\\mathrm v}\\) is luminous flux and \\(\\Omega\\) is solid angle. Luminous flux, measured in lumens (lm), expresses the rate at which radiation carries visually weighted energy. Solid angle, measured in steradians (sr), describes the angular extent of a set of directions in three-dimensional space. Consequently,\n\n\\[1\\ \\mathrm{cd}=1\\ \\mathrm{lm\\,sr^{-1}}.\\]\n\nThe differential definition matters because most real sources do not emit equally in all directions. A small cone around one direction may contain much more flux than an equally sized cone elsewhere. For a finite angular region, the quotient \\(\\Delta\\Phi_{\\mathrm v}/\\Delta\\Omega\\) gives the average luminous intensity over that region. It approximates the directional value only when the intensity varies sufficiently little within the measuring aperture. A full description is therefore generally a function \\(I_{\\mathrm v}(\\theta,\\phi)\\), with polar and azimuthal angles specifying direction relative to the source.\n\nLuminous intensity is distinct from radiant intensity, the corresponding quantity in radiometry. Radiant intensity measures emitted radiant power per solid angle in watts per steradian, without weighting it for human vision. Two sources can have the same radiant intensity but very different luminous intensities if their spectra differ. Green light near the peak of daylight visual sensitivity contributes much more to luminous intensity per watt than deep-red light; infrared radiation ordinarily contributes essentially nothing to conventional photometric quantities.\n\nA perfectly isotropic source, emitting equally in every direction, provides a simple mathematical example. Because a complete sphere subtends \\(4\\pi\\) steradians, an isotropic source of intensity \\(I_{\\mathrm v}\\) has total luminous flux\n\n\\[\\Phi_{\\mathrm v}=4\\pi I_{\\mathrm v}.\\]\n\nAn ideal isotropic one-candela source therefore emits approximately \\(12.57\\) lumens. Conversely, an isotropic source emitting \\(1{,}000\\) lumens has an intensity of about \\(79.6\\) candelas in every direction. Real lamps are rarely isotropic: supports, electrodes, housings, reflectors, and lenses alter their distributions.\n\nIn its usual application, luminous intensity characterizes a source viewed at distances large compared with its dimensions. This does not require a physically infinitesimal emitter, but it does require a geometry in which its directional output is well defined. At close distances, different parts of an extended source illuminate a receiver from appreciably different directions. A single intensity value then cannot fully describe the local light field."
    },
    {
      "heading": "Spectral weighting and the standard observer",
      "text": "The connection between luminous intensity and radiant intensity is established through a spectral weighting function. For conventional photopic measurements, representing vision at relatively high adaptation levels, the relevant function is the luminous efficiency function \\(V(\\lambda)\\). It describes the relative effectiveness of radiation of wavelength \\(\\lambda\\) in producing the standardized photometric response. If \\(I_{\\mathrm e,\\lambda}\\) denotes spectral radiant intensity per unit wavelength, then\n\n\\[I_{\\mathrm v} = K_{\\mathrm m} \\int_0^\\infty I_{\\mathrm e,\\lambda}(\\lambda)V(\\lambda)\\,\\mathrm d\\lambda,\\]\n\nwhere \\(K_{\\mathrm m}\\) is approximately \\(683\\ \\mathrm{lm\\,W^{-1}}\\). The function is normalized to unity near 555 nanometres, in the yellow-green part of the spectrum.\n\nThis weighting is not an arbitrary preference for green light. It represents experimentally established characteristics of photopic vision, in which cone photoreceptors dominate. Sensitivity falls toward both ends of the visible spectrum, so more radiant power is required at blue or deep-red wavelengths to produce the same photopic luminous intensity. For monochromatic radiation, the integral reduces to \\(I_{\\mathrm v}=K_{\\mathrm m}V(\\lambda)I_{\\mathrm e}\\). At a wavelength where \\(V(\\lambda)=0.1\\), approximately ten times as much radiant intensity is required as at the function’s maximum for the same candela value.\n\nThe International Commission on Illumination, commonly abbreviated CIE, adopted the standard photopic luminous efficiency function in 1924. Standardization made measurements comparable across laboratories and industries, rather than dependent on the eyesight of an individual observer. The function is a conventional reference, however, not a complete account of every person’s response under every viewing condition. Age, ocular transmission, field size, adaptation, and individual differences can affect actual visual experience.\n\nAt low light levels, scotopic vision is dominated by rod photoreceptors. Its standard luminous efficiency function, conventionally written \\(V'(\\lambda)\\), peaks near 507 nanometres rather than 555 nanometres. A blue-green source can therefore become relatively more visually effective as adaptation shifts toward darkness. Scotopic quantities require an explicitly identified weighting system; they should not be silently substituted for ordinary photopic candela values.\n\nWhite light also illustrates why spectrum matters. Two white sources can produce the same luminous intensity while emitting different radiant powers and having different spectral distributions. Their appearance, colour rendering, and effects on nonvisual photoreception may differ even though a photometer reports equal candelas. Luminous intensity is thus a precisely defined scalar measure of one aspect of light, not a complete spectral or biological description. Spectral measurements supply information that a single photometric value necessarily discards."
    },
    {
      "heading": "The candela and the development of standards",
      "text": "Early measurements of light relied on comparison with flames. Candles and oil or gas lamps served as reference sources, but their output depended on fuel composition, wick geometry, flame height, air supply, and operating conditions. Different countries adopted different standards, including the British standard candle and the German Hefner lamp. Such units supported practical lighting work but were difficult to reproduce consistently. The international candle, established through agreements among national laboratories in the early twentieth century, improved coordination without removing the underlying dependence on maintained source standards.\n\nA major change came with the adoption of a thermal reference based on blackbody radiation. A definition adopted internationally in 1946 and ratified by the General Conference on Weights and Measures in 1948 tied the new candle to the radiation from a blackbody at the freezing temperature of platinum. The unit received the name candela in 1948. In its later formal wording, the definition specified the luminous intensity, in the perpendicular direction, of a blackbody surface of area \\(1/600{,}000\\) square metre at platinum’s freezing temperature under a pressure of \\(101{,}325\\) pascals.\n\nThe blackbody standard replaced a chemically variable flame with a reproducible physical condition. Nevertheless, realizing it was technically demanding. Laboratories had to construct and operate high-temperature cavities, establish the freezing condition, measure an aperture accurately, and account for departures from ideal blackbody behaviour. Advances in absolute radiometry eventually made it preferable to define the unit through radiant power rather than through a particular high-temperature apparatus.\n\nIn 1979, the 16th General Conference adopted a definition based on monochromatic radiation of frequency \\(540\\times10^{12}\\) hertz. A source emitting that radiation with radiant intensity \\(1/683\\) watt per steradian has luminous intensity one candela. This frequency corresponds to a vacuum wavelength of approximately \\(555.17\\) nanometres, calculated using the speed of light. The frequency specification avoids ambiguity arising from whether wavelength is stated in vacuum or in a material medium.\n\nThe revised SI, effective from 20 May 2019, expresses the same foundation by fixing the numerical value of the luminous efficacy \\(K_{\\mathrm{cd}}\\) of monochromatic radiation at that frequency:\n\n\\[K_{\\mathrm{cd}} = 683\\ \\mathrm{lm\\,W^{-1}}.\\]\n\nThis value is exact. The candela is therefore linked to the watt and steradian through a fixed conversion at the reference frequency. Since the watt is expressible through the kilogram, metre, and second, and those units are defined using fixed constants, modern photometry connects to the wider SI system. Its distinctive feature remains the conventional visual weighting: the candela represents a standardized human visual response as well as a physical measurement of radiation."
    },
    {
      "heading": "Relationships with flux, illuminance, and luminance",
      "text": "Luminous intensity is often confused with quantities that describe different stages of light propagation. Luminous flux concerns total or partial light output. Illuminance, measured in lux, concerns luminous flux arriving per unit area of a surface. Luminance, measured in candelas per square metre, describes directional light emission or transmission per unit projected source area. These quantities are related, but none can generally replace another without additional geometric information.\n\nFor a sufficiently small source in unobstructed space, illuminance on a small receiving surface is\n\n\\[E_{\\mathrm v} = \\frac{I_{\\mathrm v}(\\theta,\\phi)\\cos\\alpha}{r^2},\\]\n\nwhere \\(r\\) is the source-to-receiver distance and \\(\\alpha\\) is the angle between the incoming light and the receiver’s normal. This combines the inverse-square law with the cosine projection of the receiving area. A source producing \\(100\\) candelas toward a normally oriented receiver produces \\(100\\) lux at one metre, \\(25\\) lux at two metres, and \\(4\\) lux at five metres, under the stated assumptions.\n\nThe distance in this relationship must be measured from an appropriate photometric reference point. Close to a large luminaire, treating the entire emitting surface as a point can cause substantial error. Obstructions, absorption, and scattering also change the received illuminance. Reflections from walls can add light that did not travel directly from source to receiver. Consequently, a room’s illuminance cannot generally be inferred from a lamp’s candela rating alone.\n\nLuminance enters when the spatial extent of the source matters. For an emitting surface of area \\(A\\), its contribution to luminous intensity in a given direction is\n\n\\[I_{\\mathrm v} = \\int_A L_{\\mathrm v}(\\mathbf{x},\\theta,\\phi)\\cos\\theta\\,\\mathrm dA,\\]\n\nwhere \\(\\theta\\) is the angle between the viewing direction and the local surface normal. A uniformly luminous planar surface viewed normally has \\(I_{\\mathrm v}=L_{\\mathrm v}A\\). If its area is \\(0.01\\ \\mathrm{m^2}\\) and its luminance is \\(1{,}000\\ \\mathrm{cd\\,m^{-2}}\\), its normal luminous intensity is \\(10\\) candelas.\n\nThese distinctions explain otherwise misleading comparisons. A small, intense light-emitting element and a large diffuse panel can have equal luminous intensity toward an observer while differing greatly in luminance and apparent size. A display is therefore commonly specified in candelas per square metre, whereas a signal lamp may be specified in candelas. A room’s lighting level is specified in lux, and a replacement lamp’s total output in lumens. The choice of quantity reflects the physical question: directional emission, surface emission, received light, or total output."
    },
    {
      "heading": "Angular distributions and optical control",
      "text": "A real source is commonly characterized by a luminous intensity distribution rather than a single candela value. This distribution records \\(I_{\\mathrm v}\\) over direction, often using one or more polar plots. For a rotationally symmetric source, intensity depends only on the polar angle from its optical axis. For a road luminaire, vehicle headlamp, or asymmetric wall-washing fixture, both angular coordinates are needed. The reference axes and orientation must be specified because the same physical device can illuminate very different regions when rotated.\n\nThe total luminous flux follows by integrating over solid angle:\n\n\\[\\Phi_{\\mathrm v} = \\int_{4\\pi} I_{\\mathrm v}(\\theta,\\phi)\\,\\mathrm d\\Omega.\\]\n\nFor rotational symmetry,\n\n\\[\\Phi_{\\mathrm v} = 2\\pi\\int_0^\\pi I_{\\mathrm v}(\\theta)\\sin\\theta\\,\\mathrm d\\theta.\\]\n\nThe factor \\(\\sin\\theta\\) is essential: equal increments of polar angle do not correspond to equal solid angles. A narrow ring of directions near the equator of a sphere covers more solid angle than one of equal angular width near a pole.\n\nAn idealized uniform conical beam illustrates the relationship between intensity and beam width. A cone of half-angle \\(\\beta\\) subtends\n\n\\[\\Omega=2\\pi(1-\\cos\\beta).\\]\n\nIf \\(1{,}000\\) lumens are distributed uniformly inside a cone of \\(10^\\circ\\) half-angle, with no light outside it, the intensity is approximately \\(10{,}500\\) candelas. The same flux distributed isotropically would yield only about \\(79.6\\) candelas. Real beams have gradual edges, sidelobes, and internal structure, so this calculation is a geometric illustration rather than a general conversion between lumens and peak candelas.\n\nA diffuse planar emitter provides another important model. If its luminance is independent of viewing direction, it is Lambertian and follows\n\n\\[I_{\\mathrm v}(\\theta)=I_{\\mathrm v}(0)\\cos\\theta\\]\n\nover the outward hemisphere. Its total flux is \\(\\pi I_{\\mathrm v}(0)\\). The declining intensity results from the shrinking projected area, not from a declining luminance. This distinction is central to understanding diffuse panels, illuminated surfaces, and many approximate models of light-emitting devices.\n\nReflectors, lenses, diffusers, and optical baffles reshape these distributions through reflection, refraction, and scattering. Concentrating light into a smaller angular region can increase peak intensity without increasing total luminous flux. Practical systems lose some flux through absorption and unwanted scattering, and passive optics are also constrained by conservation of optical étendue: they cannot arbitrarily increase a source’s radiance or luminance. Beam angle is often defined using directions at which intensity falls to half its maximum, but other conventions exist. A quoted beam angle therefore does not by itself specify either the entire distribution or the flux contained within the beam."
    },
    {
      "heading": "Measurement, calibration, and uncertainty",
      "text": "Luminous intensity is often measured indirectly from illuminance at a known distance. A calibrated photometer is positioned on a specified direction from the source, with its receiving surface normal to the incoming light. Under suitable far-field conditions,\n\n\\[I_{\\mathrm v}=E_{\\mathrm v}r^2.\\]\n\nThe experimental simplicity of this expression hides several requirements: the source must be stable, the distance must be well defined, the detector must have suitable spectral response, and stray light must be controlled. The apparatus also needs enough separation that the source can be treated according to the assumed geometry.\n\nA practical photometer usually combines a photosensitive detector with filters designed to approximate \\(V(\\lambda)\\). Its electrical output is converted to illuminance through calibration. No physical filter reproduces the standard function perfectly, so the instrument’s response depends to some extent on the source spectrum. A detector calibrated using a tungsten lamp may require a spectral mismatch correction when measuring a narrow-band or phosphor-converted light-emitting diode (LED). Sources with sharp spectral features make small response errors especially significant.\n\nA goniophotometer measures intensity as a function of direction by rotating the source, detector, or an optical arrangement between them. Different designs accommodate different source sizes and operating requirements. Some lamps must remain in a prescribed burning position because orientation affects their output. Large luminaires may require long measuring distances or near-field techniques that record spatially resolved directional emission and reconstruct the far-field distribution. Sampling must be sufficiently fine to capture narrow peaks and sharp cutoffs.\n\nPrimary photometric realizations can begin with absolute radiometric measurements. Cryogenic radiometers establish radiant power through comparison with electrical heating, and calibrated detectors transfer that scale to optical measurements. Spectral responsivity measurements and the standard visual weighting then connect radiometric quantities to photometric ones. The chain of comparisons establishes metrological traceability to the SI rather than merely agreement with a particular commercial meter.\n\nThe measurement uncertainty includes contributions from detector calibration, spectral mismatch, distance, alignment, source drift, temperature, stray light, and angular positioning. Because intensity inferred from illuminance varies as \\(r^2\\), a small relative distance uncertainty contributes approximately twice that relative uncertainty to intensity, before combination with other terms. Angular uncertainty can dominate near a sharp beam edge even when it is negligible near a broad maximum. For LEDs and electronic luminaires, stabilization time, drive current, and thermal conditions are also important because output and spectrum can change as the device warms. A defensible measurement therefore states operating conditions, geometry, and uncertainty, rather than reporting an isolated candela value."
    },
    {
      "heading": "Lighting, signalling, and product specification",
      "text": "Directional photometry is especially important in transport. A vehicle headlamp must provide useful illumination in selected parts of the road scene while limiting light in directions that could impair other road users’ vision. Its performance cannot be represented adequately by total lumens or a single peak intensity. Test procedures instead examine intensities at defined angular positions and within specified regions. The resulting distribution includes deliberately different zones for forward reach, lateral illumination, and control of upward light.\n\nSignal lights use luminous intensity for a different purpose: they must be recognizable from relevant viewing directions. Traffic signals, railway signals, navigation lights, and aviation lights have angular visibility requirements associated with their functions. Colour, background luminance, viewing distance, and atmospheric conditions all affect visibility, but directional intensity remains a central measurable characteristic. A signal may require substantial output near its intended viewing axis and much less elsewhere. Spreading the same flux over unnecessary directions can reduce useful intensity while increasing unwanted light.\n\nIn architectural and roadway lighting, intensity distributions support calculations of illuminance across surfaces. A luminaire’s photometric data can be combined with its mounting height, orientation, and position to calculate direct contributions at a grid of points. Contributions from multiple luminaires are added, while more complete models also account for reflected light. Electronic photometric files, including IES and EULUMDAT formats, communicate angular distributions for lighting design software. Some data sets express intensity in candelas per \\(1{,}000\\) lumens, allowing a normalized distribution to be scaled to a stated luminous flux.\n\nConsumer products often emphasize peak beam intensity. Flashlights and searchlights with deep reflectors can produce high candela values by concentrating light into a small angular region. This may increase the distance at which a specified on-axis illuminance is reached, even if another product emits more total lumens. From the inverse-square relationship, the ideal distance corresponding to a chosen illuminance threshold is proportional to the square root of peak intensity. Such a distance is conditional on the threshold and measurement method; it is not a universal distance at which objects become identifiable.\n\nLED specifications also demonstrate the need to read quantities carefully. Small indicator LEDs are often rated in millicandelas, with \\(1\\ \\mathrm{mcd}=10^{-3}\\ \\mathrm{cd}\\). Their stated intensity normally applies to a particular drive current and viewing direction. A narrow-angle indicator can have a higher on-axis millicandela rating than a broad-angle device with greater total luminous flux. For pulsed products, drive current, pulse duration, duty cycle, and whether the quoted value is instantaneous or averaged must also be identified. The candela rating describes optical output under specified conditions, not electrical efficiency, service life, or uniformity."
    },
    {
      "heading": "Visual interpretation and limits of the quantity",
      "text": "Luminous intensity is objective in the metrological sense: it is defined by agreed geometry, spectral weighting, and units. It is not identical to perceived brightness. Brightness is a visual attribute influenced by adaptation, contrast, apparent size, spatial arrangement, and surrounding illumination. Two sources with equal intensity can look different because one occupies a tiny angular area while the other is spread across a large surface. Their luminances and the patterns they form on the retina are different even when their far-field directional outputs are equal.\n\nThe standard photopic weighting also has limits when applied across changing adaptation conditions. At intermediate light levels, mesopic vision involves both rods and cones, and relative spectral effectiveness changes with adaptation and viewing conditions. This is relevant to outdoor lighting, where objects may be viewed against backgrounds ranging from brightly illuminated pavement to dark surroundings. Mesopic photometry provides specialized methods for particular conditions, but it does not make ordinary candela values meaningless. It identifies situations in which photopic luminous intensity alone is not a complete predictor of visual performance.\n\nTemporal behaviour introduces another limitation. A flashing source has an intensity that varies with time, so its peak, time-average, and visibility-related effective values need not coincide. For an ideal rectangular pulse train, the time-average intensity equals the pulse intensity multiplied by the duty cycle. That arithmetic average does not fully describe whether individual flashes are noticeable or how a modulation pattern is perceived. Detector integration time and electronic response can also affect readings. A measurement of modulated light must therefore distinguish the optical waveform from the particular reported summary value.\n\nPhotometric intensity is likewise not a complete measure of optical exposure. The effects of ultraviolet, visible, and infrared radiation depend on wavelength, exposure duration, source geometry, and the relevant biological mechanism. Photopic weighting was designed to represent visual effectiveness, not every biological response. Equal-candela sources can differ substantially in spectral radiant intensity, and a source with little photopic output can still emit considerable radiation outside the visible range. Radiometric and spectrally resolved quantities are needed when the physical question extends beyond conventional visual photometry.\n\nAtmospheric propagation further separates source intensity from observed appearance. Fog, rain, aerosols, and other scattering media can attenuate direct light and add a luminous veil along the line of sight. A higher source intensity does not translate into a proportionally greater practical visibility distance under every condition. In astronomy, the apparent light received from distant objects is commonly described using flux-related measures and magnitude systems rather than the candela alone. Across these applications, luminous intensity retains a precise role: it specifies how a source distributes standardized visually weighted output by direction, while additional quantities describe propagation, surfaces, time dependence, and perception."
    }
  ],
  "links": [
    "light",
    "candela",
    "international-system-of-units",
    "luminous-flux",
    "spectroscopy",
    "blackbody-radiation",
    "general-conference-on-weights-and-measures",
    "speed-of-light",
    "second",
    "illuminance",
    "scattering",
    "reflection-optics",
    "refraction",
    "calibration",
    "light-emitting-diode",
    "metrological-traceability",
    "measurement-uncertainty",
    "astronomy"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}