An aperture is an opening through which light passes in an optical system. It may be a hole in an opaque screen, an adjustable diaphragm inside a camera lens, or the usable diameter of a telescope’s light-collecting mirror. In optics, an aperture determines which rays can travel through a system and therefore influences image brightness, resolution, focus, and the appearance of out-of-focus objects.
The term has several related but distinct technical meanings. A physical opening is not necessarily the aperture that determines a system’s light-gathering capacity: lenses can magnify or reduce its apparent size. Optical design therefore distinguishes the aperture stop from its images, called the entrance and exit pupils, and from the field stop, which limits the extent of the scene. Photography commonly expresses aperture through the f-number, microscopy through numerical aperture, and astronomy through an objective’s collecting diameter. These descriptions address different aspects of the same underlying problem: controlling the spatial and angular extent of transmitted light.
Aperture stops, pupils, and ray geometry
An optical assembly may contain many openings: lens mounts, diaphragms, mirror edges, protective housings, and detector windows. The aperture stop is the element that limits the cone of rays from an on-axis object point that can pass through the system. Its identity depends on both the sizes and positions of the components. A relatively large opening can be the limiting stop if the beam is wider at its location, while a smaller opening elsewhere may transmit the entire local beam. A stop need not be a separate mechanical part; the edge of a lens or mirror can perform this function.
The entrance pupil is the image of the aperture stop formed by the optical elements preceding it, viewed from object space. If no optics precede the stop, the entrance pupil coincides with the physical opening. Otherwise, its apparent diameter and location can differ substantially from those of the diaphragm. The corresponding image formed by the elements following the stop is the exit pupil, viewed from image space. These pupil positions are essential in lens design because they determine the directions and angular spread of rays entering and leaving the system. A pupil may be real or virtual and need not lie inside the instrument.
Two characteristic rays describe this geometry. A marginal ray from an on-axis object point passes through the edge of the aperture stop and defines the boundary of the axial beam. A chief ray from an off-axis point passes through the center of the stop. Chief-ray directions help determine perspective mapping, illumination at the detector, and whether light reaches sensor pixels at favorable angles. In a telecentric system, an entrance or exit pupil lies effectively at infinity, making chief rays parallel to the optical axis on the corresponding side. Object-space telecentric lenses are useful in dimensional inspection because their magnification changes little with small changes in object distance.
The field stop has a different role: it limits the range of object positions visible through the instrument rather than primarily limiting the axial ray cone. The sensor boundary commonly determines the recorded field of a camera; a diaphragm near an intermediate image can define the field in a telescope eyepiece. Confusing field and aperture stops obscures the distinction between seeing a smaller region and collecting less light from each point within that region.
Off-axis beams may also be partially clipped by mounts or other openings, producing mechanical vignetting. Even without such clipping, image illumination can decline toward the field edge because of geometric projection and pupil effects. Thus, a quoted central aperture does not completely describe transmission across an image.
F-number, transmission, and exposure
In photography and many imaging applications, relative aperture is specified by the f-number,
where is the effective focal length and is the entrance-pupil diameter. Both lengths must use the same units. A 50 mm lens with a 25 mm entrance pupil has , conventionally written . A 200 mm lens at the same f-number has a 100 mm entrance pupil. The physical diameter alone therefore does not specify image brightness: a longer focal length distributes the image of an extended scene over a larger area.
For a distant extended subject, near the image center, and with transmission and other conditions held constant, image irradiance varies approximately as
This square relationship explains the conventional full-stop sequence , , , , , , , , and . Successive exact values differ by a factor of ; the engraved numbers are rounded. Moving one full stop toward a larger f-number halves the entrance-pupil area and approximately halves the image irradiance. Half-stop and third-stop scales divide the same logarithmic progression into smaller increments.
A photographic exposure depends on both irradiance and duration. In simplified form, radiant exposure is , where is exposure time. Closing a lens from to and doubling the exposure time approximately preserves exposure, although motion blur and depth of field change. Changing a digital camera’s ISO setting does not enlarge the aperture or increase the number of incident photons; it changes the relationship between exposure and the camera’s signal processing or output brightness.
Real lenses lose light through absorption, reflection, scattering, and other mechanisms. Cinematography therefore often uses the T-stop, a transmission-adjusted aperture number. If a lens has transmission fraction , its T-number is approximately
A lens at with has a T-number of about 2.22. Lenses set to the same calibrated T-stop are intended to produce comparable central image exposure under matching conditions. Their depth of field still depends on geometric aperture rather than the transmission correction.
The ordinary marked f-number is most directly applicable near infinity focus. At close distances, image-side ray cones generally become narrower. For a simple symmetric lens with unit pupil magnification, the effective f-number is approximately , where is the magnitude of image magnification. At life size, , this gives twice the nominal f-number and approximately four times the exposure requirement. More general lenses require a correction involving pupil magnification.
Depth of field and the appearance of defocus
An ideal geometrical lens forms a sharp image of a particular object plane at a corresponding image plane. Light from an object in front of or behind that plane converges elsewhere and forms a finite blur at the detector. Reducing aperture narrows the ray cone and reduces this defocus blur. Consequently, a greater interval of object distances may appear acceptably sharp. This interval is called depth of field, and its boundaries depend on a chosen standard of acceptable blur rather than on an abrupt physical transition.
The usual calculation introduces a circle of confusion diameter, , representing the largest permitted image-plane blur. Its value depends on sensor or film format, final enlargement, viewing distance, and the intended level of detail. An image judged sharp in a small print may reveal obvious defocus when enlarged. Pixel spacing can inform a demanding criterion, but it is not a universal replacement for a viewing-based standard: sampling, diffraction, processing, and the subject’s contrast all influence perceived sharpness.
For a thin lens under common photographic approximations, the hyperfocal distance is
When the lens is focused at this distance, objects at infinity are just within the selected blur criterion, and the near limit is approximately when is large compared with . A 50 mm lens at , using mm, has a calculated hyperfocal distance of about 10.5 m. Changing the permissible blur diameter changes that result even though the optical system remains unchanged.
Depth of field must be distinguished from depth of focus, which describes tolerance to displacement of the image plane. It must also be distinguished from the overall impression of background blur. A photograph can have a similar near-subject depth of field to another image while displaying a different background appearance because subject magnification, background distance, pupil size, and framing differ. Comparisons made at fixed camera position need not produce the same conclusions as comparisons made at fixed subject framing.
The quality of out-of-focus rendering is often called bokeh. It depends on the distribution of light within blur patches, lens aberrations, mechanical clipping, and diaphragm shape. A point light far outside the focused plane can produce a recognizable image of the pupil. Rounded diaphragm blades tend to preserve a more nearly circular outline as the aperture closes, whereas straight blades can produce polygons. Toward the frame edge, clipped pupils may create “cat’s-eye” highlights. These effects concern the character and shape of blur, not simply the amount of depth of field, and two lenses at the same f-number can render them differently.
Diffraction, aberrations, and resolution
Geometrical optics treats rays as having arbitrarily precise paths, but light also behaves as a wave. When a wave passes through a finite aperture, diffraction spreads its energy. An ideal lens therefore does not form a mathematical point image of a point source. For an unobstructed circular pupil with uniform illumination, it produces an Airy pattern: a bright central region surrounded by progressively weaker rings. The pattern follows from interference among contributions arriving through different parts of the pupil.
For an ideal circular aperture of diameter , the angular radius of the first dark ring is approximately
in radians, where is the wavelength. The Rayleigh criterion uses this angle as a conventional measure of the separation at which two point sources can be distinguished. It is not a universal boundary between resolved and unresolved objects: the answer also depends on object structure, contrast, noise, and the method used to interpret the image. Nevertheless, it expresses the central scaling that a larger aperture can resolve finer angular detail.
Near the focal plane of a lens working in air, the diameter of the Airy central disk out to its first minimum is approximately
At a wavelength of 550 nm and , this diameter is about 10.7 micrometres; at , it is about 21.5 micrometres. These values do not imply that all details smaller than the disk disappear. Image formation involves overlapping diffraction patterns, and contrast declines gradually with increasing spatial frequency. In an ideal incoherent imaging system, the cutoff spatial frequency is approximately when expressed in consistent image-plane units.
Real lenses also exhibit optical aberrations, including spherical aberration, coma, astigmatism, and chromatic aberration. Closing the aperture often suppresses contributions from outer pupil zones and reduces some of these errors. At first, stopping down can therefore improve image sharpness despite increased diffraction. At smaller openings, diffraction increasingly dominates. The aperture at which a particular lens gives its highest measured resolution depends on image position, focus, wavelength, detector sampling, and the metric used; there is no universal optimum f-number.
Aperture shape and transmission profile also matter. A central obstruction redistributes energy from the central peak into surrounding rings, while support vanes produce diffraction spikes. An apodizing filter deliberately changes pupil transmission, often softening the pupil edge to reduce sidelobes at the cost of light throughput and a broader central response. The modulation transfer function describes how strongly an imaging system reproduces contrast at different spatial frequencies and provides a more complete account than a single resolution figure.
Diaphragm mechanisms and historical development
The simplest controlled aperture is a fixed hole in an opaque plate. Such openings are mechanically stable and can be manufactured with accurately known dimensions, but changing their size requires exchanging a component or selecting another opening. In nineteenth-century photography, photographers commonly used removable stop plates. Waterhouse stops, introduced in 1858 and named after John Waterhouse, were thin plates with different-sized holes that slid into a slot in the lens barrel. They provided a repeatable means of changing relative aperture before continuously adjustable diaphragms became widespread.
The iris diaphragm uses overlapping movable blades to form a variable opening. Rotation of a control ring or movement of an actuator pivots the blades together, changing the central opening without exchanging parts. An iris is compact and continuously adjustable, but its aperture is only approximately circular. Blade number, curvature, thickness, and travel determine the opening’s shape over its adjustment range. Precision mechanisms must maintain alignment while avoiding excessive friction, backlash, and internal reflections from blade edges.
Aperture control developed alongside optical theory. Ibn al-Haytham investigated light propagation and image formation through small openings in the eleventh century. Pinhole imaging demonstrated that an opening alone could organize rays into an image without a lens. Later work on lenses, diffraction, and photographic exposure established why the opening’s size controls both illumination and sharpness. During the nineteenth and twentieth centuries, standard relative-aperture scales made exposure calculations and comparisons between lenses more consistent, although several competing numerical systems preceded the familiar f-number sequence.
Single-lens reflex cameras introduced an operational requirement: viewing and focusing are easier with the lens open, whereas the exposure may require a smaller aperture. Automatic diaphragms therefore hold the lens near maximum aperture during composition and close it to the selected setting immediately before the exposure. Earlier mechanisms used mechanical linkages between body and lens; many later systems use electronically controlled actuators. A depth-of-field preview control closes the diaphragm in advance, making the optical consequences visible, although the viewfinder image becomes darker.
Modern aperture systems balance optical accuracy against speed, sound, power consumption, and durability. Still-photography lenses commonly move between discrete commanded positions. Cinema lenses may offer smooth, unclicked adjustment so that brightness can change during a shot without visible steps or pronounced handling noise. Compact cameras and phones may use fixed openings, a small number of selectable apertures, or miniature variable diaphragms. A neutral-density filter can reduce transmitted light without narrowing the pupil, but it does not reproduce the accompanying changes in depth of field and diffraction. A shutter can also control light quantity by time rather than pupil geometry, even when its mechanism resembles an iris.
Photographic formats, sensors, and equivalent settings
Aperture comparisons between camera formats require specifying what remains constant. Cameras with different sensor sizes need different focal lengths to record the same angle of view from the same position. For a smaller format with a crop factor of 2, a 25 mm lens provides approximately the same angle of view as a 50 mm lens on the reference format. At , their entrance pupils are respectively 12.5 mm and 25 mm. Their equal f-numbers therefore describe similar image irradiance, not equal absolute pupil diameters.
With equal framing, viewpoint, shutter time, scene luminance, and transmission, equal f-numbers produce approximately equal exposure per unit sensor area. The larger sensor, however, receives more total light because its image occupies a larger area. For geometrically similar formats differing by a factor of 2 in linear dimensions, sensor areas differ by a factor of 4. In an idealized comparison with comparable detector efficiency, the larger format consequently records about four times as many photons across the whole image at the same f-number and exposure time.
This distinction connects aperture to shot noise. If the number of independently detected photons is , the standard deviation of photon-counting fluctuations is approximately , giving a signal-to-noise ratio proportional to . Four times as many detected photons therefore yields twice the photon-limited signal-to-noise ratio. Actual cameras also differ in read noise, quantum efficiency, pixel design, and processing, so sensor size and aperture alone do not determine the final noise performance.
For matched viewpoint, framing, and final display conditions, equivalent depth of field across formats is obtained approximately by scaling the f-number with crop factor. A 25 mm lens at on a format with crop factor 2 corresponds approximately to a 50 mm lens at on the reference format. Both have a 12.5 mm entrance pupil. Their total collected light can also be similar under matched shutter time and transmission, although the larger format receives lower exposure per unit area at . Output brightness then requires an appropriate sensitivity setting or processing adjustment.
Sensor architecture introduces further constraints. Microlenses direct light toward photosensitive regions, and their effectiveness can depend on ray angle. A lens with strongly oblique chief rays may interact differently with a sensor than with photographic film, particularly near image corners. Color shading and reduced efficiency may result if the sensor and lens are poorly matched. Computational background blur can imitate some visible consequences of a wide aperture by using depth estimates and image processing, but it does not physically enlarge the entrance pupil or alter the optical photon collection during that exposure.
Numerical aperture in microscopy and illumination
In microscopy, focal length divided by pupil diameter is often less useful than a measure of the angle accepted from the specimen. The standard quantity is numerical aperture,
where is the refractive index of the medium between specimen and objective, and is the half-angle of the largest accepted ray cone. A dry objective operating in air has NA below 1. Immersion oil, water, and other media permit higher numerical apertures because their refractive indices exceed that of air. High-performance oil-immersion objectives commonly reach NA values around 1.3–1.4.
Numerical aperture governs both light collection and the accessible range of spatial information. For conventional microscopy, an often-used form of the Abbe resolution estimate is
Using a wavelength of 550 nm and NA 1.4 gives about 196 nm. The exact resolution expression depends on coherence, illumination geometry, specimen properties, and the criterion applied. In transmitted-light microscopy, both the illumination numerical aperture and the objective numerical aperture can matter. In fluorescence imaging, the objective commonly supplies excitation and collects emitted light, making its acceptance angle especially important.
The condenser aperture diaphragm controls the angular distribution of illumination at the specimen. In Köhler illumination, the illumination system separates planes associated with the light source and pupils from planes associated with the specimen and image. The condenser aperture changes illumination angle, contrast, and the spatial frequencies transferred through the microscope. Closing it too far may increase apparent contrast while eliminating fine detail and introducing diffraction artifacts. The field diaphragm instead controls the illuminated specimen area and helps suppress unnecessary stray light; it is not a substitute for condenser-aperture adjustment.
In confocal microscopy, a small aperture placed in a plane conjugate to the specimen rejects much of the light originating outside the focal region. This pinhole is different from the objective aperture: it performs spatial filtering at an image plane rather than primarily setting the objective’s collection cone. A smaller confocal pinhole improves optical sectioning over part of its useful range but reduces detected signal. Its size is often expressed in Airy units, relating the physical opening to the diffraction pattern at the pinhole plane.
Similar distinctions occur in spectroscopy. An entrance slit selects a narrow spatial region and, together with the optical dispersion and imaging properties, contributes to spectral resolution. Narrowing the slit can separate nearby spectral features more clearly while reducing throughput. The pupil aperture controls a different aspect of the beam: its angular extent and the illuminated area of dispersing or focusing components. Instrument performance therefore depends on coordinated control of field openings, pupil openings, and detector geometry rather than on a single aperture dimension.
Astronomy, interferometry, and non-optical apertures
In astronomy, a telescope is commonly described by its primary aperture: the usable diameter of its objective lens or primary mirror. For an unobstructed circular collector,
A telescope with twice the diameter has four times the geometric collecting area and, at a given wavelength, approximately half the diffraction angle. These gains serve different purposes. Collecting area increases the detected signal from faint sources, whereas diameter sets the finest angular structure that an ideal filled aperture can transmit. Real throughput also depends on coatings, obscurations, detector efficiency, and losses elsewhere in the instrument.
A secondary mirror can block part of the primary aperture. For concentric circular primary and secondary diameters and , the unblocked geometric area is . The obstruction also changes the diffraction pattern rather than merely reducing the collected light. Segmented mirrors introduce gaps and pupil structure that likewise affect image formation. The James Webb Space Telescope, launched in 2021, uses a primary mirror approximately 6.5 m across made from 18 hexagonal segments. Its segmented pupil and secondary-mirror supports help produce the characteristic diffraction structure around bright point sources.
Ground-based telescope resolution is frequently limited by turbulence in the atmosphere rather than by aperture alone. Fluctuations in refractive index distort incoming wavefronts, producing astronomical seeing. Adaptive optics measures these distortions and applies rapid corrections with deformable mirrors. Its effectiveness depends on wavelength, guide-source brightness, turbulence conditions, and the number and speed of controlled modes. Without suitable correction, increasing aperture can greatly improve sensitivity while producing a smaller improvement in angular sharpness than the ideal diffraction formula predicts.
Interferometry combines signals from separated collectors. Angular resolution then depends strongly on the largest separation, or baseline, , with a characteristic scale of order . This does not make a sparse array equivalent in every respect to a filled mirror of diameter . Total collecting area remains the sum of the individual collectors’ effective areas, and the available baselines sample only particular spatial-frequency information. Aperture synthesis reconstructs images from these measurements, making array geometry and observing coverage central to the result.
The aperture concept also extends beyond visible-light imaging. In radio engineering, an antenna’s effective aperture relates incident power density to received power and is connected to antenna gain by , under the appropriate polarization and directional conventions. It need not equal the antenna’s geometric area. Synthetic-aperture radar combines observations taken along a moving platform’s path to obtain the resolving behavior of a longer antenna. In electron optics, apertures select portions of an electron beam or diffraction pattern, trading current against angular acceptance, contrast, and aberration. Across these systems, “aperture” retains its central meaning of selecting which parts of a propagating field contribute to a measurement or image.