Image analysis – Color image processing – Compression of color images
Reexamination Certificate
1999-09-14
2003-12-16
Mehta, Bhavesh M. (Department: 2625)
Image analysis
Color image processing
Compression of color images
C382S276000, C382S302000, C358S518000
Reexamination Certificate
active
06665435
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a processing method and, more particularly, to an image data processing method receiving an input image data splitted into elementary unit of information to be used in embedded applications.
The present invention also relates to an image data processing device implementing such image data processing method.
BACKGROUND OF THE INVENTION
As is well known in the technical field of image processing, during its life an image is processed by a plurality of electronic devices, that create, acquire, display store, read and write the image itself.
The image data processing device, and the corresponding processing method deal with an image acquired by means of an image acquisition device, for example a scanner.
The image data so obtained are usually organized into a raster of pixels, each pixels providing an elementary image information.
In other words, images are, at the most basic level, arrays of digital values, where a value is a collection of numbers describing the attributes of a pixel in the image. For example, in bitmaps, the above mentioned values are single binary digits.
Often, these numbers are fixed-point representation of a range of real number; for example, the integers 0 through 255 are often used to represent the numbers from 0.0 to 1.0. Often too, these numbers represent the intensity at a point of the image (gray scale) or the intensity of one color component at that point.
An important distinction has to be made in the images to be processed between achromatic and colored images.
In fact, achromatic light has only one attribute, which is the quantity of light. This attribute can be discussed in the physic sense of energy, in which case the terms intensity and luminance are used, or in the psychological sense of perceived intensity, in which case the term brightness is used.
It is useful to associate a scale with different intensity levels, for instance defining 0 as black and 1 as white; intensity levels between 0 and 1 represent different levels of grays.
The visual sensations caused by colored light are much more richer than those caused by achromatic light. Discussion on color perception usually involves three quantities, known as: hue, saturation and lightness.
1. Hue distinguishes among colors such as red, green, purple and yellow.
2. Saturation refers to how far a color is from a gray of equal intensity. Red is highly saturated; pink is relatively unsaturated; royal blue is highly saturated; sky blue is relatively unsaturated. Pastel colors are relatively unsaturated; unsaturated colors include more white light than do the vivid, saturated colors.
3. Lightness embodies the achromatic notion of perceived intensity of a reflecting object.
A fourth term, brightness, is used instead of lightness to refer to the perceived intensity of a self-luminous object (i.e. an object emitting rather than reflecting light), such as a light bulb, the sun or a CRT.
The above mentioned features of colors seem to be subjective: they depend on human observers' judgment. In reality, the branch of physics known as colorimetry provides for an objective and quantitative way of specifying colors, which can be correlated to the above perceptual classification.
A color can be represented by means of its dominant wavelength, which corresponds to the perceptual notion of hue; excitation purity corresponds to the saturation of the color; luminance is the amount or intensity of light. The excitation purity of a colored light is the proportion of pure light of the dominant wavelength and of white light needed to define the color.
A completely pure color is 100% saturated and thus contains no white light, whereas mixtures of a pure color and white light have saturations somewhere between 0 and 100%. White light and hence gray are 0% saturated, contains no color of any dominant wavelength.
Furthermore, light is fundamentally electromagnetic energy in the 400-700 nm wavelength part of the spectrum, which is perceived as the colors from violet through indigo, blue, green, yellow and orange to red. The amount of energy present at each wavelength is represented by a spectral energy distribution P(
1
), as shown in FIG.
1
.
The visual effect of any spectral distribution can be described by means of three values, i.e. the dominant wavelength, the excitation purity, and the luminance.
FIG. 2
shows the spectral distribution of
FIG. 1
, illustrating such three value. In particular, it should be noted that at the dominant wavelength there is a spike of energy of level e
2
. White light, the uniform distribution of energy level e
1
is also present.
The excitation purity depends on the relation between e
1
and e
2
: when e
1
=e
2
, excitation purity is 0%; when e
1
=0, excitation purity is 100%.
Luminance, which is proportional to the integral of the area under such curve, depends on both e
1
and e
2
.
A color model is a specification of a 3D color coordinate system and a visible subset in the coordinate system within which all colors in a particular range lie. For instance, the RGB (red, green, blue) color model is the unit cube subset of a 3D Cartesian coordinate system, as shown in FIG.
3
.
More specifically, three hardware-oriented color models are RGB, used with color CRT monitors, YIQ, i.e. the broadcast TV color system that is a re-coding of RGB transmission efficiency and for downward compatibility with black and white television and CMY (cyan, magenta, yellow) for some color-printing devices. Unfortunately none of these models are particularly easy to use because they do not relate directly to intuitive color notions of hue, saturation, and brightness. Therefore, another class of models has been developed with ease of use as a goal, such as the HSV (hue, saturation, value)—sometimes called HSB (hue, saturation, brightness), HLS (hue, lightness, saturation) and HVC (hue, value, chroma) models.
With each model is also given a means of converting to some other specification.
As stated above, the RGB color model used in color CRT monitors and color raster graphics employs a Cartesian coordinate system. The RGB primaries are additive primaries; that is the individual contributions of each primary are added together to yield the result. The main diagonal of the cube, with equal amounts of each primary, represents the gray levels: black is (0,0,0); white is (1,1,1).
Following such gray line implies the change of the three Cartesian value R, G and B at the same time, as shown with a point-dotted line in
FIG. 4A
; this situation weights the computational charge of the image processing steps requiring the individuation of gray regions.
The RGB model is hardware-oriented. By contrast HSV (as well as HSB or HLC) model is user-oriented, being based on the intuitive appeal of the artist's tint, shade, and tone. The coordinate system is cylindrical, as shown in FIG.
4
B.
The HSV model (like the HLC model) is easy to use. The grays all have S=0 and they can be removed from an image data raster by means of a cylindrical filter in proximity of the V axes, as shown in
FIG. 5
; moreover, the maximally saturated hues are at S=1, L=0.5.
The HLS color model is a reduced model obtained from the HSV cylindrical model, as shown in
FIG. 6
; the reduction of the color space is due to the fact that some colors cannot be saturated. Such space subset is defined is a hexcone or six-sided pyramid, as shown in FIG.
7
. The top of the hexcone corresponds to V=1 which contains the relatively bright colors. The colors of the V=1 plane are not all of the same perceived brightness however.
Hue or H, is measured by the angle around the vertical axis with red at 0° green at 120° and so on (see FIG.
7
), Complementary colors in the HSV hexcone are 180° opposite one another. The value of S is a ratio ranging from 0 on the center line (V axis) to 1 on the triangular sides of the hexcone.
The hexcone is one unit high in V, with the apex at the origin. The point at the apex is black and has a V coordinate of 0. At thi
Caldato Claudio
Heins Douglas
Monaci Andrea
Bayat Ali
Mehta Bhavesh M.
LandOfFree
Image data processing method and corresponding device does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image data processing method and corresponding device, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image data processing method and corresponding device will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3164029