|
|
| Line 1: |
Line 1: |
| {{about|ringing artifacts in signal processing, particularly image processing|ringing in electronics and signals generally|ringing (signal)}}
| |
| [[File:Ringing artifact example.png|thumb|240px|Image showing ringing artifacts.
| |
|
| |
|
| 3 levels on each side of transition: overshoot, first ring, and (faint) second ring.]]
| |
| [[File:Ringing artifact example - original.png|thumb|240px|Same image without ringing artifacts.]]
| |
|
| |
|
| In [[signal processing]], particularly [[digital image processing]], '''ringing artifacts''' are [[Artifact (error)|artifacts]] that appear as spurious signals near sharp transitions in a signal. Visually, they appear as bands or "ghosts" near edges; audibly, they appear as "echos" near [[Transient (acoustics)|transients]], particularly sounds from [[percussion instrument]]s; most noticeable are the [[pre-echo]]s. The term "ringing" is because the output signal oscillates at a fading rate around a sharp transition in the input, similar to a [[Bell (instrument)|bell]] after being struck. As with other artifacts, their minimization is a criterion in [[filter design]].
| | It depends on the quality of the Wordpress theme but even if it's not a professional one you will be able to average 50-60$ EACH link. What I advise you do next is save the backup data file to a remote place like a CD-ROM, external disk drive if you have one or a provider such as Dropbox. This is a service where people write articles using a specific keyword you have given them. Donor oocytes and menopausal pregnancy: Oocyte donation to women of advanced reproductive age: pregnancy results and obstetrical outcomes in patients 45 years and older. provided by Word - Press Automatic Upgrade, so whenever you need to update the new version does not, it automatically creates no webmaster. <br><br>Right starting from social media support to search engine optimization, such plugins are easily available within the Word - Press open source platform. You may either choose to link only to the top-level category pages or the ones that contain information about your products and services. It allows Word - Press users to easily use HTML5 the element enable native video playback within the browser. Furthermore, with the launch of Windows 7 Phone is the smart phone market nascent App. Once you've installed the program you can quickly begin by adding content and editing it with features such as bullet pointing, text alignment and effects without having to do all the coding yourself. <br><br>If you treasured this article and you simply would like to receive more info concerning [http://ll.my/wordpressbackup978215 backup plugin] nicely visit the web-page. ve labored so hard to publish and put up on their website. Now if we talk about them one by one then -wordpress blog customization means customization of your blog such as installation of wordpress on your server by wordpress developer which will help you to acquire the SEO friendly blog application integrated with your site design as well as separate blog administration panel for starting up your own business blog,which demands a experienced wordpress designer. After age 35, 18% of pregnancies will end in miscarriage. You can allow visitors to post comments, or you can even allow your visitors to register and create their own personal blogs. Article Source: Stevens works in Internet and Network Marketing. <br><br>The primary differences are in the plugins that I install, as all sites don't need all the normal plugins. High Quality Services: These companies help you in creating high quality Word - Press websites. One of the great features of Wordpress is its ability to integrate SEO into your site. It's now become a great place to sell it thanks to Woo - Commerce. The Pakistani culture is in demand of a main surgical treatment. <br><br>Website security has become a major concern among individuals all over the world. Mahatma Gandhi is known as one of the most prominent personalities and symbols of peace, non-violence and freedom. By the time you get the Gallery Word - Press Themes, the first thing that you should know is on how to install it. Page speed is an important factor in ranking, especially with Google. I have never seen a plugin with such a massive array of features, this does everything that platinum SEO and All In One SEO, also throws in the functionality found within SEO Smart Links and a number of other plugins it is essentially the swiss army knife of Word - Press plugins. |
| | |
| == Introduction ==
| |
| [[File:High accuracy settling time measurements figure 1.png|thumb|The main cause of ringing artifacts is [[overshoot (signal)|overshoot]] and oscillations in the [[step response]] of a filter.]]
| |
| The main cause of ringing artifacts is due to a signal being [[bandlimited]] (specifically, not having high frequencies) or passed through a [[low-pass filter]]; this is the [[frequency domain]] description.
| |
| In terms of the [[time domain]], the cause of this type of ringing is the ripples in the [[sinc function]],<ref name="Bankman">{{citation
| |
| |title=Handbook of medical imaging
| |
| |first=Isaac N.
| |
| |last=Bankman
| |
| |publisher=Academic Press
| |
| |year=2000
| |
| |isbn=978-0-12-077790-7
| |
| |url=http://books.google.com/?id=UHkkPBnhT-MC
| |
| }}, section I.6, Enhancement: Frequency Domain Techniques, [http://books.google.com/books?id=UHkkPBnhT-MC&pg=RA3-PA16#PRA3-PA16,M1 p. 16]</ref> which is the [[impulse response]] (time domain representation) of a perfect low-pass filter. Mathematically, this is called the [[Gibbs phenomenon]].
| |
| | |
| One may distinguish [[overshoot (signal)|overshoot]] (and undershoot), which occurs when transitions are accentuated – the output is higher than the input – from ringing, where ''after'' an overshoot, the signal overcorrects and is now below the target value; these phenomena often occur together, and are thus often conflated and jointly referred to as "ringing".
| |
| | |
| The term "ringing" is most often used for ripples in the ''time'' domain, though it is also sometimes used for ''frequency'' domain effects:<ref name="Chitode">[http://books.google.com/books?id=LbUFjehqSdwC Digital Signal Processing], by J.S.Chitode, Technical Publications, 2008, ISBN 978-81-8431-346-8, [http://books.google.com/books?id=LbUFjehqSdwC&pg=RA3-PA1-IA104#PRA3-PA1-IA103,M1 4 - 70]</ref>
| |
| windowing a filter in the time domain by a rectangular function causes ripples in the ''frequency'' domain for the same reason as a brick-wall low pass filter (rectangular function in the ''frequency'' domain) causes ripples in the ''time'' domain, in each case the Fourier transform of the rectangular function being the sinc function.
| |
| | |
| There are related artifacts caused by other [[frequency domain]] effects,
| |
| and [[#Similar phenomena|similar artifacts]] due to unrelated causes.
| |
| | |
| == Causes ==
| |
| === Description ===
| |
| [[File:Sinc function (normalized).svg|thumb|The [[sinc function]], the [[impulse response]] for an ideal [[low-pass filter]], illustrating ringing for an impulse.]]
| |
| [[File:Gibbs phenomenon 10.svg|thumb|The [[Gibbs phenomenon]], illustrating ringing for a [[step function]].]]
| |
| By definition, ringing occurs when a non-oscillating input yields an oscillating output: formally, when an input signal which is [[monotonic function|monotonic]] on an interval has output response which is not monotonic. This occurs most severely when the [[impulse response]] or [[step response]] of a [[Filter (signal processing)|filter]] has oscillations – less formally, if for a spike input, respectively a step input (a sharp transition), the output has bumps. Ringing most commonly refers to step ringing, and that will be the focus.
| |
| | |
| Ringing is closely related to [[overshoot (signal)|overshoot]] and undershoot, which is when the output takes on values higher than the maximum (respectively, lower than the minimum) input value: one can have one without the other, but in important cases, such as a [[low-pass filter]], one first has overshoot, then the response bounces back below the steady-state level, causing the first ring, and then oscillates back and forth above and below the steady-state level. Thus overshoot is the first step of the phenomenon, while ringing is the second and subsequent steps. Due to this close connection, the terms are often conflated, with "ringing" referring to both the initial overshoot and the subsequent rings.
| |
| | |
| If one has a [[linear time invariant]] (LTI) filter, then one can understand the filter and ringing in terms of the impulse response (the time domain view), or in terms of its Fourier transform, the [[frequency response]] (the frequency domain view). Ringing is a ''time'' domain artifact, and in [[filter design]] is traded off with desired frequency domain characteristics: the desired frequency response may cause ringing, while reducing or eliminating ringing may worsen the frequency response.
| |
| | |
| === sinc filter ===
| |
| {{main|sinc filter}} <!-- lower-case intentional, as this is a function name -->
| |
| [[File:Sine integral.svg|thumb|The [[Sine integral]] for positive values, exhibiting oscillation.]]
| |
| The central example, and often what is meant by "ringing artifacts", is the ideal ([[Brick-wall filter|brick-wall]]) [[low-pass filter]], the [[sinc filter]]. This has an oscillatory impulse response function, as illustrated above, and the step response – its integral, the [[Sine integral]] – thus also features oscillations, as illustrated at right.
| |
| | |
| These ringing artifacts are not results of imperfect implementation or windowing: the ideal low-pass filter, while possessing the desired frequency response, necessarily causes ringing artifacts in the ''time'' domain.
| |
| | |
| === Time domain ===
| |
| In terms of impulse response, the correspondence between these artifacts and the behavior of the function is as follows:
| |
| * impulse undershoot is equivalent to the impulse response having negative values,
| |
| * impulse ringing (ringing near a point) is precisely equivalent to the impulse response having oscillations, which is equivalent to the derivative of the impulse response alternating between negative and positive values,
| |
| * and there is no notion of impulse overshoot, as the unit impulse is assumed to have infinite height (and integral 1 – a [[Dirac delta function]]), and thus cannot be overshot.
| |
| | |
| Turning to step response,
| |
| the step response is the integral of the [[impulse response]]; formally, the value of the step response at time ''a'' is the integral <math>\int_{-\infty}^a</math> of the impulse response. Thus values of the step response can be understood in terms of ''tail'' integrals of the impulse response.
| |
| | |
| Assume that the overall integral of the impulse response is 1, so it sends constant input to the same constant as output – otherwise the filter has [[gain]], and scaling by gain gives an integral of 1.
| |
| * Step undershoot is equivalent to a tail integral being negative, in which case the magnitude of the undershoot is the value of the tail integral.
| |
| * Step overshoot is equivalent to a tail integral being greater than 1, in which case the magnitude of the overshoot is the amount by which the tail integral exceeds 1 – or equivalently the value of the tail in the other direction, <math>\int_a^\infty,</math> since these add up to 1.
| |
| * Step ringing is equivalent to tail integrals alternating between increasing and decreasing – taking derivatives, this is equivalent to the impulse response alternating between positive and negative values.<ref>{{citation
| |
| |title=Principles of Digital Image Synthesis
| |
| |url=http://books.google.com/?id=6KUsFm7L-LQC
| |
| |first=Andrew S
| |
| |last=Glassner
| |
| |edition=2
| |
| |publisher=Morgan Kaufmann
| |
| |year=2004
| |
| |isbn =978-1-55860-276-2
| |
| | |
| }}, [http://books.google.com/books?id=6KUsFm7L-LQC&pg=PA518#PPA518,M1 p. 518]</ref> Regions where an impulse response are below or above the ''x''-axis (formally, regions between zeros) are called '''lobes,''' and the magnitude of an oscillation (from peak to trough) equals the integral of the corresponding lobe.
| |
| | |
| The impulse response may have many negative lobes, and thus many oscillations, each yielding a ring, though these decay for practical filters, and thus one generally only sees a few rings, with the first generally being most pronounced.
| |
| | |
| Note that if the impulse response has small negative lobes and larger positive lobes, then it will exhibit ringing but not undershoot or overshoot: the tail integral will always be between 0 and 1, but will oscillate down at each negative lobe. However, in the sinc filter, the lobes monotonically decrease in magnitude and alternate in sign, as in the [[alternating harmonic series]], and thus tail integrals alternate in sign as well, so it exhibits overshoot as well as ringing.
| |
| | |
| Conversely, if the impulse response is always nonnegative, so it has no negative lobes – the function is a [[probability distribution]] – then the step response will exhibit neither ringing nor overshoot or undershoot – it will be a monotonic function growing from 0 to 1, like a [[cumulative distribution function]]. Thus the basic solution from the time domain perspective is to use filters with nonnegative impulse response.
| |
| | |
| === Frequency domain ===
| |
| The frequency domain perspective is that ringing is caused by the sharp cut-off in the rectangular passband in the frequency domain, and thus is reduced by smoother [[roll-off]], as discussed below.<ref name="Bankman" /><ref name="MIP">[http://books.google.com/books?id=uGWmR0f_350C Microscope Image Processing], by Qiang Wu, Fatima Merchant, Kenneth Castleman, ISBN 978-0-12-372578-3 [http://books.google.com/books?id=uGWmR0f_350C&pg=RA1-PA71 p. 71]</ref>
| |
| | |
| == Solutions ==
| |
| Solutions depend on the parameters of the problem: if the cause is a low-pass filter, one may choose a different filter design, which reduces artifacts at the expense of worse frequency domain performance. On the other hand, if the cause is a band-limited signal, as in JPEG, one cannot simply replace a filter, and ringing artifacts may prove hard to fix – they are present in [[JPEG 2000]] and many audio compression codecs (in the form of [[pre-echo]]), as discussed in the [[#Examples|examples]].
| |
| | |
| === Low-pass filter ===
| |
| [[File:DisNormal01.svg|thumb|The [[Gaussian function]] is non-negative and non-oscillating, hence causes no overshoot or ringing.]]
| |
| | |
| If the cause is the use of a brick-wall low-pass filter, one may replace the filter with one that reduces the time domain artifacts, at the cost of frequency domain performance. This can be analyzed from the time domain or frequency domain perspective.
| |
| | |
| In the time domain, the cause is an impulse response that oscillates, assuming negative values. This can be resolved by using a filter whose impulse response is non-negative and does not oscillate, but shares desired traits. For example, for a low-pass filter, the [[Gaussian filter]] is non-negative and non-oscillatory, hence causes no ringing. However, it is not as good as a low-pass filter: it rolls off in the passband, and leaks in the stopband: in image terms, a Gaussian filter "blurs" the signal, which reflects the attenuation of desired higher frequency signals in the passband.
| |
| | |
| A general solution is to use a [[window function]] on the sinc filter, which cuts off or reduces the negative lobes: these respectively eliminate and reduce overshoot and ringing. Note that truncating some but not all of the lobes eliminates the ringing beyond that point, but does not reduce the amplitude of the ringing that is not truncated (because this is determined by the size of the lobe), and increases the magnitude of the overshoot if the last non-cut lobe is negative, since the magnitude of the overshoot is the integral of the ''tail,'' which is no longer canceled by positive lobes.
| |
| | |
| Further, in practical implementations one at least truncates sinc, otherwise one must use infinitely many data points (or rather, all points of the signal) to compute every point of the output – truncation corresponds to a rectangular window, and makes the filter practically implementable, but the frequency response is no longer perfect.<ref>{{Harv|Allen|Mills|2004}} Section 9.3.1.1 Ideal Filters: Low pass, [http://books.google.com/books?id=ZmyKvXQmQwIC&pg=PA621 p. 621]</ref>
| |
| In fact, if one takes a brick wall low-pass filter (sinc in time domain, rectangular in frequency domain) and truncates it (multiplies with a rectangular function in the time domain), this convolves the frequency domain with sinc (Fourier transform of the rectangular function) and causes ringing in the ''frequency'' domain,<ref name="Chitode" /> which is referred to as ''[[ripple (electrical)#Frequency-domain ripple|ripple]].'' In symbols, <math>\mathcal{F}(\mathrm{sinc}\cdot \mathrm{rect}) = \mathrm{rect} * \mathrm{sinc}.</math> The frequency ringing in the stopband is also referred to as [[side lobe]]s. Flat response in the passband is desirable, so one windows with functions whose Fourier transform has fewer oscillations, so the frequency domain behavior is better.
| |
| | |
| Multiplication in the time domain corresponds to convolution in the frequency domain, so multiplying a filter by a window function corresponds to convolving the Fourier transform of the original filter by the Fourier transform of the window, which has a smoothing effect – thus windowing in the time domain corresponds to smoothing in the frequency domain, and reduces or eliminates overshoot and ringing.<ref>{{Harv|Allen|Mills|2004}} [http://books.google.com/books?id=ZmyKvXQmQwIC&pg=PA623 p. 623]</ref>
| |
| | |
| In the [[frequency domain]], the cause can be interpreted as due to the sharp (brick-wall) cut-off, and ringing reduced by using a filter with smoother roll-off.<ref name="Bankman" /> This is the case for the Gaussian filter, whose magnitude [[Bode plot]] is a downward opening parabola (quadratic roll-off), as its Fourier transform is again a Gaussian, hence (up to scale) <math>e^{-x^2}</math> – taking logarithms yields <math>-x^2.</math>
| |
| | |
| {{external media
| |
| |image1=[http://books.google.com/books?id=dunqt1rt4sAC&pg=RA4-PA332 Butterworth filter impulse response and frequency response graphs]<ref name="OpAmp">[http://books.google.com/books?id=dunqt1rt4sAC Op Amp applications handbook],
| |
| by Walter G. Jung, Newnes, 2004, | |
| ISBN 978-0-7506-7844-5, [http://books.google.com/books?id=dunqt1rt4sAC&pg=RA4-PA332 p. 332]</ref>
| |
| }}
| |
| In [[electronic filter]]s, the trade-off between frequency domain response and time domain ringing artifacts is well-illustrated by the [[Butterworth filter]]: the frequency response of a Butterworth filter slopes down linearly on the log scale, with a first-order filter having slope of −6 [[decibel|dB]] per [[octave]], a second-order filter –12 dB per octave, and an ''n''th order filter having slope of <math>-6n</math> dB per octave – in the limit, this approaches a brick-wall filter. Thus, among these the, first-order filter rolls off slowest, and hence exhibits the fewest time domain artifacts, but leaks the most in the stopband, while as order increases, the leakage decreases, but artifacts increase.<ref name="MIP" />
| |
| | |
| == Benefits ==
| |
| [[File:Accutance.svg|thumb|Artificially added overshoot around the left bar increases [[acutance]].]]
| |
| While ringing artifacts are generally considered undesirable, the initial overshoot (haloing) at transitions increases [[acutance]] (apparent sharpness) by increasing the derivative across the transition, and thus can be considered as an enhancement.<ref name="mitchell">{{cite conference
| |
| |title=Reconstruction filters in computer-graphics
| |
| |first=Don P.
| |
| |last=Mitchell
| |
| |coauthors=Netravali, Arun N.
| |
| |url=http://www.mentallandscape.com/Papers_siggraph88.pdf
| |
| |doi=10.1145/54852.378514
| |
| |conference=ACM SIGGRAPH International Conference on Computer Graphics and Interactive Techniques
| |
| |conferenceurl=http://portal.acm.org/toc.cfm?id=54852&type=proceeding&coll=GUIDE&dl=GUIDE,ACM&CFID=30538218&CFTOKEN=95411512
| |
| |pages=221–228
| |
| |date=August 1988
| |
| |volume=22
| |
| |number=4
| |
| |ISBN=0-89791-275-6
| |
| }}</ref>
| |
| | |
| == Related phenomena ==
| |
| === Overshoot ===
| |
| [[File:Sinc function (normalized).svg|thumb|The sinc function has negative tail integrals, hence has overshoot.]]
| |
| [[File:Lanczos-kernel.svg|thumb|The Lanczos 2-lobed filter exhibits only overshoot, while the 3-lobed filter exhibits overshoot and ringing.]]
| |
| {{main|Overshoot (signal)}}
| |
| Another artifact is [[overshoot (signal)|overshoot]] (and undershoot), which manifests itself not as rings, but as an increased jump at the transition. It is related to ringing, and often occurs in combination with it.
| |
| | |
| Overshoot and undershoot are caused by a negative tail – in the sinc, the integral from the first zero to infinity, including the first negative lobe. While ringing is caused by a following ''positive'' tail – in sinc, the integral from the second zero to infinity, including the first non-central positive lobe.
| |
| Thus overshoot is ''necessary'' for ringing,{{Dubious|date=July 2013}} but can occur separately: for example, the 2-lobed [[Lanczos filter]] has only a single negative lobe on each side, with no following positive lobe, and thus exhibits overshoot but no ringing, while the 3-lobed Lanczos filter exhibits both overshoot and ringing, though the windowing reduces this compared to the sinc filter or the truncated sinc filter.
| |
| | |
| Similarly, the convolution kernel used in [[bicubic interpolation]] is similar to a 2-lobe windowed sinc, taking on negative values, and thus produces overshoot artifacts, which appear as halos at transitions.
| |
| | |
| === Clipping ===
| |
| {{main|Clipping (audio)}}
| |
| Following from overshoot and undershoot is [[Clipping (audio)|clipping]].
| |
| If the signal is bounded, for instance an 8-bit or 16-bit integer, this overshoot and undershoot can exceed the range of permissible values, thus causing clipping.
| |
| | |
| Strictly speaking, the clipping is caused by the combination of overshoot and limited numerical accuracy, but it is closely associated with ringing, and often occurs in combination with it.
| |
| | |
| Clipping can also occur for unrelated reasons, from a signal simply exceeding the range of a channel.
| |
| | |
| === Ringing and ripple ===
| |
| {{main|Ringing (signal)|Ripple (filters)}}
| |
| [[File:Tchebyscheff5.png|thumb|Frequency response of a 5th order [[Chebyshev filter]], exhibiting [[ripple (electrical)#Frequency-domain ripple|ripple]].]]
| |
| | |
| In signal processing and related fields, the general phenomenon of time domain oscillation is called '''[[Ringing (signal)|ringing]],''' while frequency domain oscillations are generally called '''[[ripple (electrical)#Frequency-domain ripple|ripple]],''' though generally not "rippling".
| |
| | |
| A key source of ripple in digital signal processing is the use of [[window function]]s: if one takes an [[infinite impulse response]] (IIR) filter, such as the sinc filter, and windows it to make it have [[finite impulse response]], as in the [[window design method]], then the frequency response of the resulting filter is the convolution of the frequency response of the IIR filter with the frequency response of the window function. Notably, the frequency response of the rectangular filter is the sinc function (the rectangular function and the sinc function are [[Conjugate variables|Fourier dual]] to each other), and thus truncation of a filter in the time domain corresponds to multiplication by the rectangular filter, thus convolution by the sinc filter in the frequency domain, causing ripple. In symbols, the frequency response of <math>\mathrm{rect}(t) \cdot h(t)</math> is <math>\mathrm{sinc}(t) * \hat h(t).</math> In particular, truncating the sinc function itself yields <math>\mathrm{rect}(t) \cdot \mathrm{sinc}(t)</math> in the time domain, and <math>\mathrm{sinc}(t) * \mathrm{rect}(t)</math> in the frequency domain, so just as low-pass filtering (truncating in the frequency domain) causes ''ringing'' in the time domain, truncating in the time domain (windowing by a rectangular filter) causes ''ripple'' in the frequency domain.
| |
| | |
| == Examples ==
| |
| === JPEG ===
| |
| [[File:Asterisk with jpg-artefacts.png|thumb|Extreme example of JPEG artifacts, including ringing: cyan (= white minus red) rings around a red star.]]
| |
| [[File:Dctjpeg.png|thumb|[[Discrete cosine transform]] basis functions.]]
| |
| [[JPEG]] compression can introduce ringing artifacts at sharp transitions, which are particularly visible in text.
| |
| | |
| This is a due to loss of high frequency components, as in step response ringing.
| |
| [[JPEG#Block_splitting|JPEG uses 8×8 blocks]], on which the [[discrete cosine transform]] (DCT) is performed. The DCT is a [[Fourier-related transform]], and ringing occurs because of loss of high frequency components or loss of precision in high frequency components.
| |
| | |
| They can also occur at the edge of an image: since JPEG splits images into 8×8 blocks, if an image is not an integer number of blocks, the edge cannot easily be encoded, and solutions such as filling with a black border create a sharp transition in the source, hence ringing artifacts in the encoded image.
| |
| | |
| Ringing also occurs in the [[wavelet]]-based [[JPEG 2000]].
| |
| | |
| JPEG and JPEG 2000 have other artifacts, as illustrated above, such as blocking ("[[jaggies]]") and edge busyness ("[[mosquito noise]]"), though these are due to specifics of the formats, and are not ringing as discussed here.
| |
| | |
| Some illustrations:
| |
| * [http://www.stat.columbia.edu/~jakulin/jpeg/artifacts.htm Baseline JPEG and JPEG2000 Artifacts Illustrated]
| |
| {| class="wikitable" style="text-align:center"
| |
| ! Image !! Lossless compression !! Lossy compression
| |
| |-
| |
| ! Original
| |
| | [[File:Lossless-circle.png|100px]] || [[File:Lossy-circle.jpg|100px]]
| |
| |-
| |
| ! Processed by<br/>[[Canny edge detector]],<br/>highlighting artifacts.
| |
| | [[File:Lossless-circle-canny.png|100px]] || [[File:Lossy-circle-canny.png|100px]]
| |
| |}
| |
| | |
| === Pre-echo ===
| |
| [[File:2006-07-06 Crash Zildjian 14.jpg|thumb|[[Pre-echo]] occurs in percussions such as cymbals.]]
| |
| {{main|Pre-echo}}
| |
| In [[audio signal processing]], ringing can causes echoes to occur before and after [[transient (acoustics)|transients]], such as the impulsive sound from [[percussion instrument]]s, such as [[cymbal]]s (this is ''impulse'' ringing). The ([[Causal filter|causal]]) echo after the transient is not heard, because it is masked by the
| |
| transient, an effect called [[temporal masking]]. Thus only the ([[Anti-causal filter|anti-causal]]) echo before the transient is heard, and the phenomenon is called [[pre-echo]].
| |
| | |
| This phenomenon occurs as a [[compression artifact]] in audio compression algorithms that use [[Fourier-related transforms]], such as [[MP3]], [[Advanced Audio Coding|AAC]], and [[Vorbis]].
| |
| | |
| == Similar phenomena ==
| |
| Other phenomena have similar symptoms to ringing, but are otherwise distinct in their causes. In cases where these cause circular artifacts around point sources, these may be referred to as "rings" due to the round shape (formally, an [[Annulus (mathematics)|annulus]]), which is unrelated to the "ringing" (oscillatory decay) frequency phenomenon discussed on this page.
| |
| | |
| === Edge enhancement ===
| |
| {{main|Edge enhancement}}
| |
| | |
| [[Edge enhancement]], which aims to increase edges, may cause ringing phenomena, particularly under repeated application, such as by a DVD player followed by a television. This may be done by ''high''-pass filtering, rather than low-pass filtering.<ref name="MIP" />
| |
| | |
| === Special functions ===
| |
| [[File:Airy-pattern.svg|thumb|The [[Airy pattern]], caused by [[Fraunhofer diffraction]].]]
| |
| Many [[special functions]] exhibit oscillatory decay, and thus convolving with such a function yields ringing in the output; one may consider these ringing, or restrict the term to unintended artifacts in frequency domain signal processing.
| |
| | |
| [[Fraunhofer diffraction]] yields the [[Airy disk]] as [[point spread function]], which has a ringing pattern.
| |
| | |
| [[File:Bessel Functions (1st Kind, n=0,1,2).svg|thumb|A few [[Bessel function]]s of the first kind, showing oscillatory decay.]]
| |
| The [[Bessel function]] of the first kind, <math>J_0,</math> which is related to the [[Airy function]], exhibits such decay.
| |
| | |
| [[File:Spherical-aberration-disk.jpg|thumb|Combinations of [[defocus]] and [[spherical aberration]] exhibit ring artifacts.]]
| |
| In cameras, a combination of [[defocus]] and [[spherical aberration]] can yield circular artifacts ("ring" patterns). However, the pattern of these artifacts need not be similar to ringing (as discussed on this page) – they may exhibit oscillatory decay (circles of decreasing intensity), or other intensity patterns, such as a single bright band.
| |
| | |
| === Interference ===
| |
| [[Ghosting (television)|Ghosting]] is a form of [[television interference]] where an image in repeated, though this is not ringing; it can be interpreted as convolution with a function which is 1 at the origin and ε (the intensity of the ghost) at some distance, which is formally similar to the above functions (a single discrete peak, rather than continuous oscillation).
| |
| | |
| === Lens flare ===
| |
| {{details|Lens flare}}
| |
| | |
| [[File:Lens Flare.jpg|thumb|[[Lens flare]].]]
| |
| In photography, [[lens flare]] is a defect where various circles can appear around highlights, and with ghosts throughout a photo, due to undesired light, such as reflection and scattering off elements in the lens.
| |
| | |
| === Visual illusions ===
| |
| [[File:Mach band.svg|thumb|[[Mach bands]]]]
| |
| Visual illusions can occur at transitions, as in [[Mach bands]], which perceptually exhibit a similar undershoot/overshoot to the Gibbs phenomenon.
| |
| | |
| == See also ==
| |
| {{colbegin}}
| |
| * [[Artifact (error)]]
| |
| * [[Digital artifact]]
| |
| * [[sinc filter]]<!-- intentionally lowercase, as a function name -->
| |
| * [[Brick-wall filter]]
| |
| * [[Chromatic aberration]]
| |
| * [[Ghosting (television)]]
| |
| * [[Gibbs phenomenon]]
| |
| * [[Low-pass filter]]
| |
| * [[Pre-echo]]
| |
| * [[Purple fringing]]
| |
| {{colend}}
| |
| | |
| == References ==
| |
| {{reflist}}
| |
| * {{citation
| |
| |title=Signal analysis: time, frequency, scale, and structure
| |
| |first1=Ronald L.
| |
| |last1=Allen
| |
| |first2=Duncan W.
| |
| |last2=Mills
| |
| |publisher=Wiley-IEEE
| |
| |year=2004
| |
| |url=http://books.google.com/?id=ZmyKvXQmQwIC
| |
| |isbn=978-0-471-23441-8
| |
| }}
| |
| | |
| [[Category:Signal processing]]
| |
| [[Category:Computer graphic artifacts]]
| |