Hubbry Logo
Quantization (image processing)Quantization (image processing)Main
Open search
Quantization (image processing)
Community hub
Quantization (image processing)
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Quantization (image processing)
Quantization (image processing)
Quantization (image processing)
View on Wikipedia
from Wikipedia

Quantization, involved in image processing, is a lossy compression technique achieved by compressing a range of values to a single quantum (discrete) value. When the number of discrete symbols in a given stream is reduced, the stream becomes more compressible. For example, reducing the number of colors required to represent a digital image makes it possible to reduce its file size. Specific applications include DCT data quantization in JPEG and DWT data quantization in JPEG 2000.

Color quantization

[edit]
Main article: Color quantization

Color quantization reduces the number of colors used in an image; this is important for displaying images on devices that support a limited number of colors and for efficiently compressing certain kinds of images. Most bitmap editors and many operating systems have built-in support for color quantization. Popular modern color quantization algorithms include the nearest color algorithm (for fixed palettes), the median cut algorithm, and an algorithm based on octrees.

It is common to combine color quantization with dithering to create an impression of a larger number of colors and eliminate banding artifacts.

Grayscale quantization

[edit]

Grayscale quantization, also known as gray level quantization, is a process in digital image processing that involves reducing the number of unique intensity levels (shades of gray) in an image while preserving its essential visual information. This technique is commonly used for simplifying images, reducing storage requirements, and facilitating processing operations. In grayscale quantization, an image with N intensity levels is converted into an image with a reduced number of levels, typically L levels, where L<N. The process involves mapping each pixel's original intensity value to one of the new intensity levels. One of the simplest methods of grayscale quantization is uniform quantization, where the intensity range is divided into equal intervals, and each interval is represented by a single intensity value. Let's say we have an image with intensity levels ranging from 0 to 255 (8-bit grayscale). If we want to quantize it to 4 levels, the intervals would be [0-63], [64-127], [128-191], and [192-255]. Each interval would be represented by the midpoint intensity value, resulting in intensity levels of 31, 95, 159, and 223 respectively.

The formula for uniform quantization is:

Where:

  • Q(x) is the quantized intensity value.
  • x is the original intensity value.
  • Δ is the size of each quantization interval.

Let's quantize an original intensity value of 147 to 3 intensity levels.

Original intensity value: x=147

Desired intensity levels: L=3

We first need to calculate the size of each quantization interval:

Using the uniform quantization formula:

Rounding 191.25 to the nearest integer, we get

So, the quantized intensity value of 147 to 3 levels is 191.

Frequency quantization for image compression

[edit]

The human eye is fairly good at seeing small differences in brightness over a relatively large area, but not so good at distinguishing the exact strength of a high frequency (rapidly varying) brightness variation. This fact allows one to reduce the amount of information required by ignoring the high frequency components. This is done by simply dividing each component in the frequency domain by a constant for that component, and then rounding to the nearest integer. This is the main lossy operation in the whole process. As a result of this, it is typically the case that many of the higher frequency components are rounded to zero, and many of the rest become small positive or negative numbers.

As human vision is also more sensitive to luminance than chrominance, further compression can be obtained by working in a non-RGB color space which separates the two (e.g., YCbCr), and quantizing the channels separately.[1]

Quantization matrices

[edit]

A typical video codec works by breaking the picture into discrete blocks (8×8 pixels in the case of MPEG[1]). These blocks can then be subjected to discrete cosine transform (DCT) to calculate the frequency components, both horizontally and vertically.[1] The resulting block (the same size as the original block) is then pre-multiplied by the quantization scale code and divided element-wise by the quantization matrix, and rounding each resultant element. The quantization matrix is designed to provide more resolution to more perceivable frequency components over less perceivable components (usually lower frequencies over high frequencies) in addition to transforming as many components to 0, which can be encoded with greatest efficiency. Many video encoders (such as DivX, Xvid, and 3ivx) and compression standards (such as MPEG-2 and H.264/AVC) allow custom matrices to be used. The extent of the reduction may be varied by changing the quantizer scale code, taking up much less bandwidth than a full quantizer matrix.[1]

This is an example of DCT coefficient matrix:

A common quantization matrix is:

Dividing the DCT coefficient matrix element-wise with this quantization matrix, and rounding to integers results in:

For example, using −415 (the DC coefficient) and rounding to the nearest integer

Typically this process will result in matrices with values primarily in the upper left (low frequency) corner. By using a zig-zag ordering to group the non-zero entries and run length encoding, the quantized matrix can be much more efficiently stored than the non-quantized version.[1]

See also

[edit]

References

[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Quantization (image processing)

View on Grokipedia
from Grokipedia
Quantization in image processing is a fundamental technique for converting continuous or high-precision pixel intensity values into a finite set of discrete levels, effectively reducing the bit depth of an image while introducing some loss of information. This process maps input amplitudes within defined decision intervals to corresponding reconstruction levels, enabling the representation of images with fewer bits per pixel, such as scaling from 12-16 bits to 8 bits for standard display compatibility. As a form of lossy data compression, it groups intervals of luminance or color data into single quantum values, minimizing storage and transmission requirements without preserving every original detail. The quantization process typically occurs during analog-to-digital (A/D) conversion, where a continuous signal, such as , is transformed into discrete digital counts via a that relates input to output levels. For uniform quantization, the step size remains constant across the range, calculated as b=fmax⁡−fmin⁡2m−1b = \frac{f_{\max} - f_{\min}}{2^m - 1}, where mm is the number of bits and 2m2^m defines the total levels (e.g., 256 levels for 8 bits). Non-uniform quantization, in contrast, adjusts level spacing based on signal statistics or human perception, such as logarithmic scaling to allocate more precision to darker tones, thereby reducing visible artifacts. Key types include scalar quantization, which operates on individual pixel values to reduce gray levels; vector quantization, which processes groups of pixels as multidimensional vectors for more efficient compression; and color quantization, which applies to RGB or other color spaces to limit the palette while preserving visual fidelity, often using algorithms like median cut. These methods are essential in applications such as image acquisition, where they handle sensor outputs; transmission and compression standards like ; and display adaptation for devices with limited , including and . However, quantization introduces error, modeled as with variance approximately σn2=b212\sigma_n^2 = \frac{b^2}{12} for uniform distributions, potentially causing contouring effects when levels drop below 64 (6 bits) per . Techniques like dithering or the Lloyd-Max quantizer, which optimizes levels for , help mitigate these distortions. Overall, the improves by about 6 dB per additional bit, balancing quality against computational efficiency.

Fundamentals

Definition and Process

Quantization in image processing refers to the process of mapping a large set of continuous or high-precision input values, such as pixel intensities, to a smaller set of discrete output levels, thereby reducing the precision of the data while aiming to preserve essential visual information. This many-to-one mapping function typically decreases the number of bits required to represent each pixel—for instance, reducing 8-bit grayscale values (256 levels) to 4-bit values (16 levels)—to minimize storage requirements or facilitate efficient transmission over bandwidth-limited channels. As a lossy technique inherent to digital representation, quantization introduces irreversible information loss but is crucial for practical digital imaging systems. The general process of quantization begins with analog-to-digital conversion, where a continuous analog signal is first sampled spatially to form a discrete grid of , followed by the assignment of discrete intensity levels to those pixel values. An input signal xx is divided into quantization intervals defined by a step size Δ\Delta, and each value within an interval is mapped to the nearest representative level, producing the quantized output q(x)q(x). For uniform quantization, this is mathematically expressed as q(x)=Δ⋅\roundxΔ,q(x) = \Delta \cdot \round{\frac{x}{\Delta}}, where \round⋅\round{\cdot} denotes the rounding operation to the nearest integer, ensuring even spacing of levels across the input range. This step size Δ\Delta is often determined by the total range of input values divided by the number of desired levels minus one, balancing data reduction against perceptual quality degradation. The origins of quantization in image processing trace back to the early , when advancements in and the prompted the development of techniques for efficient digital representation of images in resource-constrained environments. Pioneering efforts at NASA's and Bell Laboratories applied quantization during the processing of lunar and planetary images; for example, the 1964 Ranger 7 mission transmitted the first digital images from the , requiring quantization to convert analog camera signals into manageable digital formats for transmission back to over limited bandwidth. These early applications, building on foundational concepts from the and , established quantization as a core prerequisite for , assuming basic representation while addressing the challenges of converting real-world analog scenes into discrete data.

Uniform and Non-Uniform Quantization

Uniform quantization divides the input signal range into equal intervals, assigning each interval to a discrete level with a constant step size Δ\Delta. This approach simplifies and , making it suitable for hardware and software processing in systems. However, it assumes uniform perceptual importance across the , which often leads to higher visible errors in regions where the human visual system is more sensitive, such as low-intensity areas. The quantization error for uniform quantization, modeled as uniformly distributed noise over [−Δ/2,Δ/2][-\Delta/2, \Delta/2], has a (MSE) approximated by \MSE≈Δ212\MSE \approx \frac{\Delta^2}{12}. Non-uniform quantization employs variable step sizes to allocate more levels to signal regions with higher perceptual , reducing overall for a fixed number of bits. A common example is the logarithmic μ\mu-law , which compresses the nonlinearly before uniform quantization and expands it afterward. The μ\mu-law function is given by y=\sign(x)⋅ln⁡(1+μ∣x∣/max⁡)ln⁡(1+μ)y = \sign(x) \cdot \frac{\ln(1 + \mu |x| / \max)}{\ln(1 + \mu)}, where μ\mu is the compression (typically 255 for standard applications) and max⁡\max is the peak signal value; the companded signal yy is then uniformly quantized. This method has been adapted for image processing to enhance in restoration tasks, improving signal-to-noise ratios at low input levels. To compare distortion between schemes, the (SQNR) for uniform quantization of a full-scale sinusoidal input with bb bits is \SQNR=6.02b+1.76\SQNR = 6.02b + 1.76 dB, assuming quantization uniformly distributed over the Nyquist bandwidth. Non-uniform schemes can achieve higher effective SQNR in perceptually relevant regions by adapting steps to the signal's probability density, though they increase complexity. Perceptual considerations drive the adoption of non-uniform quantization, as the human visual system exhibits greater sensitivity to changes in than in , with significantly lower sensitivity to color differences, allowing for coarser quantization steps for chrominance components to minimize visible artifacts without sacrificing efficiency, as implemented in standards like through tailored quantization tables derived from psychovisual experiments.

Spatial Domain Quantization

Grayscale Quantization

Grayscale quantization reduces the bit depth of single-channel intensity images, mapping continuous or high-precision values to a of discrete levels to minimize storage and transmission requirements while aiming to preserve visual fidelity. In an 8-bit grayscale image, intensities range from 0 to 255, representing 256 possible levels; quantization might truncate or round these to fewer levels, such as 16 in a 4-bit representation, by dividing the range into uniform intervals and assigning each pixel to the nearest reproduction value. For instance, with 16 levels spaced at intervals of approximately 17 (0, 17, 34, ..., 255), a value of 200 would be rounded to 204, the nearest level, effectively compressing the but potentially introducing visible steps in smooth gradients. Key techniques for grayscale quantization include posterization, which applies hard thresholds to enforce abrupt transitions between discrete intensity bands, creating a stylized, banded appearance often used for artistic or simplified rendering, and bit-plane slicing, which decomposes the into eight binary planes (from least to most significant bit) to selectively discard lower planes for precision loss. In posterization, thresholds are set to map ranges of intensities to fixed output levels, such as reducing a 256-level to 8 levels by grouping every 32 values into one, resulting in distinct tonal regions without intermediate shades. Bit-plane slicing, conversely, represents each pixel's value as a binary sum across planes (e.g., 204 = 11001100 in binary, contributing to planes 2, 3, 5, 6, 7, and 8); retaining only higher planes (e.g., the top 5) achieves 5-bit quantization (32 levels) with high compression ratios, as lower planes often contain or fine details that can be omitted with minimal perceptual impact. The perceptual quality of quantized images is commonly evaluated using the (PSNR), which quantifies distortion relative to the original by comparing mean squared differences in intensities. Defined as \PSNR=10log⁡10\MAX2\MSE,\PSNR = 10 \log_{10} \frac{\MAX^2}{\MSE}, where \MAX\MAX is the maximum possible value (typically 255 for 8-bit ) and \MSE\MSE is the between the original and quantized images, higher PSNR values indicate better ; for example, reducing from 8-bit to 4-bit often yields PSNRs around 30-40 dB for natural images, balancing compression against visible artifacts like . This metric, while objective, correlates reasonably with human for distortions but is sensitive to uniform quantization schemes. Applications of grayscale quantization span early monochrome displays, where hardware limitations necessitated reducing intensity levels to achieve feasible resolutions, and , where it enables bandwidth savings without compromising diagnostic utility. In early systems, such as those using conventional cards limited to 8-bit output, techniques like video across RGB channels of color monitors effectively extended grayscale depth to 12 bits (over 4,000 levels) by combining signals for smoother gradients in vision research and displays. In medical contexts, quantizing 12-16 bit images to 10-bit levels supports transmission over constrained networks while aligning with human visual discrimination of approximately 700-900 just-noticeable differences, ensuring perceptual linearity via standards like GSDF for efficient storage and remote diagnostics.

Color Quantization

Color quantization is the process of reducing the number of distinct colors in a multi-channel , typically from millions to hundreds, to facilitate storage, transmission, or display on devices with limited while aiming to minimize perceptual distortion. Unlike scalar quantization in single-channel images, color quantization treats pixels as vectors in a multi-dimensional , requiring algorithms that cluster colors to form a representative palette. This approach is essential for applications such as palette-based formats (e.g., or with reduced colors) and real-time rendering on constrained hardware. Color space selection plays a critical role in achieving visually faithful results. In the RGB color space, which is device-dependent and not perceptually uniform, equal Euclidean distances do not correspond to equal perceived color differences, potentially leading to noticeable banding in quantized images. Perceptually uniform spaces like CIELAB (Lab*) address this by approximating human vision, where the L* component represents lightness and a*, b* capture opponent colors, ensuring that quantization errors are more evenly distributed across perceived color variations. Similarly, separates (Y) from (Cb, Cr), allowing independent quantization; this is advantageous because the human visual system exhibits higher spatial sensitivity to luminance changes than to chrominance, enabling coarser quantization of chroma channels without substantial quality loss. Quantization in such spaces often involves transforming the image, quantizing channels separately or jointly, and inverse-transforming back to RGB for display. Several have been developed for palette-based color quantization, focusing on partitioning the to select representative colors. The median-cut , introduced by Heckbert, operates by recursively subdividing the RGB into hyper-rectangular "boxes" (voxels) based on color population. The process begins by representing all unique colors as points in the 24-bit RGB cube; the box containing the most colors is selected, its longest dimension (R, G, or B axis with the greatest range) is identified, and the box is split perpendicular to that axis at the median color count to balance populations on both sides. This splitting continues until the desired number of boxes (e.g., 256) is reached, with each final box's or average color serving as a palette entry. Median-cut is computationally efficient and produces palettes that approximate uniform color distribution but can struggle with sparse color regions. The quantization method, proposed by Gervautz and Purgathofer, employs a to cluster colors, offering advantages in handling variable color densities. Colors are inserted into an where each level divides the RGB cube into eight equal subcubes (octants) based on binary splits along each axis (e.g., R > 128 or ≤128). Nodes represent color subcubes, with leaf nodes storing color counts; the tree is built to a fixed depth (typically 8 levels for 256 leaves) or until all colors are isolated. To generate a palette of size N, the is pruned by repeatedly removing the least populous nodes and promoting their parent as a representative color, weighted by subtree populations. This approach excels at preserving rare colors and is faster for building than exhaustive clustering, though it may require post-processing for optimal palette quality. Palette generation typically reduces a 24-bit color space (16,777,216 possible colors) to 256 or fewer entries, after which each original is mapped to the nearest palette color using a distance metric. Error minimization occurs via nearest-neighbor assignment, often computed with in the chosen ; for perceptual accuracy, distances are preferably calculated in Lab* to align with human sensitivity. This mapping can introduce quantization error, measured as the average distortion between original and quantized colors, but the palette ensures compact representation suitable for indexed-color formats. Vector quantization provides a more general framework for color palette creation, treating each RGB pixel as a three-dimensional vector and learning a codebook of prototype vectors (palette colors) to minimize overall distortion. The seminal Linde-Buzo-Gray (LBG) algorithm, a variant of k-means clustering, initializes a codebook with k randomly selected or subsampled colors, then iteratively assigns each image color vector to the nearest codebook vector (using ) and updates codevectors as the centroids of their assigned clusters until convergence. In RGB space, serves as the distortion measure, though transformations to or are common to incorporate perceptual weighting, reducing the impact of errors in less-sensitive channels. This method yields high-quality palettes for arbitrary k but can be computationally intensive, with convergence depending on initialization to avoid local minima.

Frequency Domain Quantization

Principles in Compression

In transform-based schemes, quantization plays a pivotal role by reducing the precision of transform coefficients after applying transforms such as the (DCT) or , thereby discarding less perceptually important frequency components to achieve data reduction. In standards like , this lossy step follows the DCT on 8x8 blocks of the image, where coefficients representing high-frequency details are often quantized to zero, effectively eliminating fine spatial variations that contribute minimally to overall image fidelity. Similarly, in JPEG2000, quantization applied to wavelet subbands targets high-frequency components across scales, leveraging the multi-resolution nature of the transform to prioritize low-frequency energy. The human visual system (HVS) exhibits reduced sensitivity to high spatial frequencies, as characterized by its contrast sensitivity function, which peaks at low to mid frequencies and declines sharply beyond approximately 60 cycles per degree. This perceptual property, rooted in Weber's law—where the in stimulus intensity is proportional to the stimulus magnitude—allows for coarser quantization of (AC) coefficients, which capture high-frequency details, while (DC) coefficients, representing average intensity, receive finer quantization to preserve structure. The quantization process typically involves dividing each transform by a scalar or matrix-derived value and to the nearest , formalized as Q(u,v)=\roundF(u,v)q(u,v)Q(u,v) = \round{\frac{F(u,v)}{q(u,v)}}, where F(u,v)F(u,v) denotes the transform at frequency indices (u,v)(u,v), and q(u,v)q(u,v) is the quantizer value. Larger q(u,v)q(u,v) values for high frequencies result in many to zero, concentrating in fewer non-zero terms for efficient . Due to its irreversible nature, quantization introduces information loss proportional to the degree of bit reduction, yet this enables high compression ratios, such as 10:1 in for visually acceptable quality in color images. The loss is controlled by adjusting quantizer values, balancing file size against perceptual distortion while exploiting HVS limitations to minimize visible artifacts.

Quantization Matrices and Tables

In quantization for , quantization matrices are predefined 8x8 tables applied to (DCT) coefficient blocks in standards like , where each entry scales the quantization step for a specific frequency component. These matrices ensure that the 64 DCT coefficients within an 8x8 spatial block are divided by corresponding table values before rounding, reducing precision while prioritizing perceptual quality. For () components, the baseline matrix features smaller values in the upper-left corner, corresponding to low-frequency coefficients that are more visible to the , allowing finer quantization steps there. A standard example matrix, recommended in the specification for typical viewing conditions, is shown below:
1611101624405161
1212141926586055
1413162440576956
1417222951878062
182237566810910377
243555648110411392
49647887103121120101
7292959811210010399
The design of these matrices draws from psycho-visual models, which quantify human sensitivity to frequency variations, assigning larger quantization steps (higher table values) to high-frequency components that are less perceptible, thereby minimizing visible distortion while achieving compression. This approach balances reduction with image fidelity, as low-frequency changes impact perceived quality more significantly than high-frequency ones. Adaptive methods adjust these matrices to control compression quality, commonly through a scaling factor applied uniformly across the table. In JPEG implementations like , a quality factor QQ ranging from 1 (highest compression, lowest ) to 100 (lowest compression, highest ) determines the scaling ss, yielding the adjusted matrix via q′(u,v)=q(u,v)⋅sq'(u,v) = q(u,v) \cdot s, where q(u,v)q(u,v) is the base entry at frequency indices (u,v)(u,v). (libjpeg documentation on quality scaling) This scaling inversely affects file size and distortion: lower QQ increases ss, coarsening quantization for greater compression. JPEG baseline standards specify separate matrices for and channels to exploit differing human visual sensitivities, with the matrix using finer steps (e.g., values from 10 to 99) compared to (e.g., values often 99 for high frequencies, reflecting reduced acuity for color details). A typical matrix example is:
1718244799999999
1821266699999999
2426569999999999
4766999999999999
9999999999999999
9999999999999999
9999999999999999
9999999999999999
Extensions in (ISO/IEC 15444-1) replace fixed matrices with adaptive scalar quantization thresholds applied per subband, using uniform step sizes Δ\Delta that can vary by component and resolution level for more flexible rate-distortion control. Unlike JPEG's DCT-based matrices, these thresholds enable irreversible compression with psycho-visually tuned steps, often smaller for low-frequency subbands akin to priorities.

Artifacts and Mitigation

Common Visual Artifacts

Quantization in image processing introduces several visible distortions, collectively known as artifacts, which degrade perceptual quality by altering smooth transitions and structural details. Common artifacts include contour banding, posterization, blocking, and mosquito noise. Contour banding manifests as false contours or step-like bands in smooth gradients, particularly in low-texture regions such as skies or skin tones, where continuous intensity variations are mapped to discrete levels. Posterization refers to the reduction of color gradations into flat, distinct regions, resulting in a stylized, painted appearance that lacks subtlety in shaded areas. Blocking appears as grid-like discontinuities at block boundaries in compressed images, while mosquito noise presents as flickering, high-frequency noise around sharp edges, often resembling insect-like halos. These artifacts arise primarily from the process inherent to quantization. In the spatial domain, uniform quantization divides the intensity range into equal intervals, leading to visible steps in low-contrast areas where small intensity differences exceed the quantization step size, exacerbating banding and posterization in smooth regions. For instance, reducing bit depth from 8 bits (256 levels) to 3 bits (8 levels) per channel creates pronounced steps, as seen in images where gradients appear as abrupt jumps rather than seamless transitions. In the , coarse quantization of transform coefficients, such as in or MPEG, truncates high-frequency components, causing blocking due to independent processing of pixel blocks and introducing quantization that manifests as mosquito noise around edges through of high-frequency details. The severity of these artifacts is often assessed using perceptual metrics that go beyond traditional (MSE), which inadequately captures human visual . The Structural Similarity Index (SSIM) provides a more suitable measure than MSE by evaluating , contrast, and structural distortions, offering better correlation with subjective quality judgments for some quantization-induced impairments like blocking, though it has limitations for banding. Factors influencing artifact visibility include the extent of bit depth reduction—severe posterization occurs at 2-3 bits per channel, limiting colors to 4-8 levels—and viewing conditions, such as high-resolution displays that amplify the perception of fine steps in low-contrast regions.

Dithering and Error Diffusion

Dithering techniques mitigate the visibility of quantization artifacts by intentionally introducing controlled into the prior to quantization, thereby decorrelating the quantization and preventing the formation of regular patterns such as or banding. This approach trades distribution for a more perceptually uniform appearance, where the added mimics natural variations that the human is less sensitive to. Dithering can be classified into random and ordered types. Random dithering adds uncorrelated pseudo-random noise, typically uniform white noise, to each value before quantization, which effectively randomizes error placement but can sometimes introduce excessive graininess. Ordered dithering, in contrast, employs a fixed threshold matrix tiled across the image; a seminal example is the Bayer matrix, a dispersed-dot generated recursively from smaller matrices, where each entry determines the threshold for deciding quantization based on the input intensity relative to the normalized matrix value. Error diffusion represents a class of adaptive spatial dithering algorithms that achieve superior perceptual quality by propagating quantization errors to neighboring s rather than adding independent . The Floyd-Steinberg algorithm, introduced in , processes the in raster order: for each , the input intensity is modified by accumulated errors from previously processed neighbors, quantized to the nearest available level, and the resulting error is distributed to unprocessed adjacent s using a specific weighting mask—typically 7/16 to the right neighbor, 3/16 to the below-left, 5/16 to the below, and 1/16 to the below-right. The modified input intensity for (i,j)(i, j) is given by ei,j=Ii,j+∑k,lwk,l⋅erri−k,j−l,e_{i,j} = I_{i,j} + \sum_{k,l} w_{k,l} \cdot \mathrm{err}_{i-k,j-l}, where Ii,jI_{i,j} is the original intensity, wk,lw_{k,l} are the diffusion weights, and erri−k,j−l\mathrm{err}_{i-k,j-l} are the quantization errors from prior pixels. This process shapes the error spectrum to reduce low-frequency components, enhancing detail preservation. These methods find widespread application in halftoning for digital printers, where binary ink dots simulate continuous grayscales through patterned dot placement, as in error diffusion's ability to produce sharp edges without moiré patterns. In low-bit-depth displays, dithering reduces visible banding in gradients, improving apparent smoothness. For color images, it offers perceptual benefits by distributing errors across channels, making color shifts less detectable to the eye. Variants of dithering address limitations in pattern visibility and adaptability. Blue-noise dithering generates error patterns with power concentrated in mid-to-high spatial frequencies, avoiding the clustered dots of low-pass noise or the harshness of white noise, resulting in less perceptible artifacts; this is achieved through void-and-cluster algorithms or minimized variance metrics on the point process. Adaptive threshold variants, such as variable-coefficient , dynamically adjust diffusion weights or thresholds based on local image statistics to further suppress worming artifacts or enhance edge rendition.

References

  1. https://www.sciencedirect.com/topics/engineering/image-quantization
  2. https://www.princeton.edu/~cuff/ele201/kulkarni_text/digitizn.pdf
  3. https://www.cis.rit.edu/class/simg361/Notes_090910.pdf
  4. https://www.cl72.org/090imagePLib/books/Gonzales%2CWoods-Digital.Image.Processing.4th.Edition.pdf
  5. https://www.sciencedirect.com/topics/engineering/uniform-quantization
  6. https://www.nasa.gov/history/55-years-ago-ranger-7-photographs-the-moon/
  7. https://www.cis.rit.edu/class/simg712-90/notes/14-Quantization.pdf
  8. https://apps.dtic.mil/sti/tr/pdf/ADA486862.pdf
  9. https://www.analog.com/media/en/training-seminars/tutorials/MT-001.pdf
  10. https://ijg.org/files/Wallace.JPEG.pdf
  11. https://www.stsci.edu/stsci/resources/imaging/papers/rector2007/files/Rector2005.pdf
  12. https://ijarcce.com/upload/2017/february-17/IJARCCE%2072.pdf
  13. https://www.ni.com/en/shop/data-acquisition-and-control/add-ons-for-data-acquisition-and-control/what-is-vision-development-module/peak-signal-to-noise-ratio-as-an-image-quality-metric.html
  14. https://www.sciencedirect.com/science/article/pii/S0165027003001742
  15. https://pmc.ncbi.nlm.nih.gov/articles/PMC3043920/
  16. https://engineering.purdue.edu/~ee538/Image_compression_wavelets_jpeg2000.pdf
  17. https://web.stanford.edu/class/ee368b/Handouts/09-HumanPerception.pdf
  18. https://cs.stanford.edu/people/eroberts/courses/soco/projects/data-compression/lossy/jpeg/coeff.htm
  19. https://www.w3.org/Graphics/JPEG/itu-t81.pdf
  20. http://www.cs.northwestern.edu/~jet/docs/2005OuyangTumblinQuant.pdf
  21. https://www.cl.cam.ac.uk/~rkm38/pdfs/denes2019banding_model.pdf
  22. https://ieeexplore.ieee.org/document/4025970
  23. https://doi.org/10.1117/12.387203
  24. https://ieeexplore.ieee.org/document/1284395
  25. https://cv.ulichney.com/papers/1988-blue-noise.pdf
  26. https://perso.liris.cnrs.fr/victor.ostromoukhov/publications/pdf/SIGGRAPH01_varcoeffED.pdf
Add your contribution
Related Hubs
User Avatar
No comments yet.