AU681641B2 - Method for processing an image in a computerized system - Google Patents

Abstract

Computerized processing method for an image, wherein the original image is sampled in order to retain only a grid of points of reduced or much lower definition than that of the original image and corresponding to the image displayed on a monitor. The processing to be carried out is broken down into elementary operations which may be summed at any time to display the partial result on the screen. The operations are carried out only on the functions and, it is only the global function of the intermediary stage, or of the whole treatment, which is then calculated with functional interpolation for the different intermediary points of the grid used, according to the definition of the final image to be obtained, and it is only after having calculated the function thus obtained for all the intermediary points that the data corresponding to said function are calculated. The process applies in general to image processing for the production of films, particularly color films for printing.

Description

P \OPERKAT\V4O33.94.147 -121697 -1- This invention concerns a procedure for image processing in a computerised system.

Such processing involves, in the widest sense, all processing operations that can be carried out on an original image. These operations cover modifications to the color or appearance of all or part of an original image, color changes such as might be achieved using an airbrush, or importation into the original image of other images, which may themselves be processed either in combination with or separately from the original image.

A range of image processing procedures exists already. What these procedures all have in common is the fact that each modification to the image involves recalculation operations in order to rearrange the dots of the original image (or those resulting from the most recent modification) into a new arrangement. Such rearrangement operations are irreversible.

In addition (and this is the greatest disadvantage of known procedures) the image held in memory offers considerably better definition than that displayed on the operator's monitor screen. Because the final image to be obtained (usually for printing purposes) is based on the 15 one held in memory, it in turn will have a far higher definition than that which it is possible to reproduce on the screen. The operator therefore can never view the results of his or her work as they will appear in reality.

~All the calculations arising from processing carried out on the original image, including importation of external images, are calculations carried out on the pixels of the 20 original image, conditioned by the definition of the original image or of the image to be obtained.

Again, operations carried out on an image usually require a high degree of processing power. If this processing power is not available, then the time required to carry out the operation becomes unacceptably long. Moreover, the scope and sophistication of the oeoe•: 25 operations which it is possible to carry out on the images is reduced. For example, airbrush strokes are extremely limited in width, precisely because of the processing power needed to calculate the changes in the image concerned.

The irreversible nature of image processing using current procedures is a particularly serious drawback, for the operator is prevented from applying any second thoughts. The only way of correcting an airbrush stroke which does not achieve the desired effect is to "1L~ P \OPERKAT\&6038-94,147. 12169 superimpose a new stroke (instead of simply to erase the unsuccessful one).

This is all the more serious in that checking an image displayed on a screen is in general not as reliable or precise as checking the proof image outputted by the system if only because of the extremely reduced definition of the monitor image by comparison with that of the final desired result.

To sum up, current computerised image processing in order to obtain a high definition image for the preparing of print film suffers from a dual disadvantage: it requires large amounts of processing power, and changes made by the operator are irreversible.

In embodiments of the invention described below, on the other hand, it is intended to provide a computerised image processing procedure, enabling the operator to carry out advanced graphic operations, rapidly, with the ability to reverse decisions as required without in any way affecting the definition or precision of the final image.

In one aspect, the invention provides a method of processing a digital image in a computerised system in order to obtain an image to be used for display or printing, in such 15 a way that the checking of each elementary operation is effected by displaying the result of :..this operation on a monitor screen, the procedure being characterised by the following: the original image to be processed is sampled in accordance with a definition grid so as to retain from all the dots contained within the original image only a number of dots equal at the most to the number that can be displayed on the monitor 20 screen, the processing of the image is broken down into elementary recurrent operations each broken down in turn into three parts and providing, based on the result of the previous elementary operation (layer a new result (layer these three parts added to each other representing: adopting the new layer of a color dot from the previous layer (i-1) with a weighting ranging from 0 to 100% (positive or negative value), importing an external imago, (Ii) into the layer i, by importing of a color dot from the image after chromatic and geometric transformation of this dot to add it to the color dot of the layer the degree of replacement of the dot of 3, Q the layer by the dot imported from the image (li) being defined by a scalar (Pi(x,y)) *i 1, v j\ K tk 9 -R Ca- bBt -~-PIA P \OPERKAT\64038-4.147 1216/97 -3with values from 0 to 100% (positive or negative values), carrying out a chromatic modification on the dot of the layer each elementary operation being defined by the equation taking account of the previous operation p,(xy) a,(xy) p,(xy) y,(xy) in which: a,(x,y) o o sc r r a o o o p

O

p.(x,y) Pi(x,y) Ii Pi(x,y) is a scalar function of the dot corresponding to the presence at this dot of the image resulting from the previous elementary operation is a function representing the previous elementary operation, is a scalar function corresponding to the presence at dot of a dot corresponding to the imported image, represents the imported image made up of a set of dots, is the function of image import representing the chromatic geometric transfer of one of the set of dots in the image towards the layer to which is applied the elementary operation Ii[Pi(x,y)] is the function corresponding to the import of the image, y,(x,y) is a chromatic function representing a color transformation function carried out on a dot each of the terms iPi[Pi(x,y)] and yi(x,y) not being precluded from being nil while the term ai(x,y) is never nil for all the dots j i -i- P \OI'ERKAT4038-94.147. 1216/91 -4- The elementary operations may be effected to obtain a function representing the i first elementary operations in order to obtain a function whose parameters are defined at all the dots of the definition grid jqS(xy) I y(xy) q number of imported images, in this global function: aj(x,y) is a scalar analogous to the scalar a(x,y) of a elementary function Ij represents an image j to import P(x,y) is an import function analogous to the previous import functions Pi(x,y) 15 y(x,y) is a chromatic function analogous to chromatic functions yi(x,y), the global function being defined by interpolating it at the intermediate dots between the dots of the definition grid, these intermediate dots depending on the definition required for the final image, the pixels being calculated for each dot to be obtained.

The invention also provides apparatus for editing, in a digital image processing S. system, an original image comprising: display means for displaying image portions; means for generating a first image portion at a user-specified resolution, including: 1) means for generating and storing an image-pyramid data structure encoding the original image and a plurality of reduced-resolution subimages derived from said encoded original image; P \OI'PER\AT64O38-94 147- 12/6 97 S

S

means for structuring the encoded original image and the encoded reduced-resolution subimages within said image-pyramid data structure as a plurality of individually-accessible tiles; means for accepting an image-resolution specification entered by a user, the image-resolution specification defining an image resolution within a range of image resolutions spanned by the image-pyramid data structure; and means for selecting and accessing one or more tiles of said encoded original image or reduced-resolution subimages to generate data encoding the first image portion at an image resolution corresponding to the image resolution defined by the image-resolution specification; means for capturing image-editing commands entered by a user for editing said first image portion by modifying the first image portion corresponding to imposing the image-editing effects specified by the image-editing commands on the first image portion, each image-editing command being representative of one of a plurality of editing modes; means for generating editing-effect parameters that define the modifications to said first image portion corresponding to imposing the image-editing effects specified by the image-editing commands; means for storing said editing-effect parameters in an editing-effect data structure representative of a plurality of layers, each layer being associated with an editing mode corresponding to the editing mode associated with the editing-effect parameters stored in the layer; and means for generating data encoding a second image portion by combining said editing-effect parameters in the editing-effect data structure with data encoding said first image portion to impose the image-editing effects defined by said parameters on the first image portion so that in operation said second image portion may be displayed on the display means to display the first image portion as modified by image-editing effects specified by the image-editing commands.

The invention also provides a method for editing an original image, in a digital imaging processing system, comprising the steps of:

HA,

P \OPER\KAT& 38-94.147 12/6/97 -6generating and displaying a first image portion at a user-specified resolution, including the steps of: 1) generating and storing an image pyramid data structure encoding the original image and a plurality of reduced-resolution subimages derived from said encoded original image; structuring the encoded original image and the encoded reducedresolution subimages within said image-pyramid data structure as a plurality of individually-accessible tiles; accepting an image-resolution specification entered by a user, the image-resolution specification defining an image resolution within a range of image resolutions spanned by the image-pyramid data structure; selecting and accessing one or more tiles of said encoded original image or reduced-resolution subimages to generate data encoding the first image portion at an image resolution corresponding to the image resolution 15 defined by the image-resolution specification; and 5) displaying the first image portion so generated; capturing image-editing commands entered by a user for editing said first image portion by modifying the first image portion corresponding to imposing the image-editing effects specified by the image-editing commands on the first image portion, each image-editing command being representative of one of a plurality of editing modes; generating editing-effect parameters that define the modifications to said first image portion corresponding to imposing the image-editing effects specified by the image-editing commands; storing said editing-effect parameters in an editing-effect data structure representative of a plurality of layers, each layer being associated with an editing mode corresponding to the editing mode associated with the editing-effect parameters stored in the layer; generating data encoding a second image portion by combining said editingeffect parameters in the editing-effect data structure with data encoding said first P \()PER\KAT\6438-94. 147- 12/6197 -7image portion to impose the image-editing effects defined by said parameters on the first image portion; and displaying said second image portion to display the first image portion as modified by image-editing effects specified by the image-editing commands.

The invention also provides apparatus for editing an original image in a digital image processing system comprising: display means for displaying image portions; means for generating a first image portion at a user-specified resolution, including: 1) means for generating and storing an image-pyramid data structure encoding the original image and a plurality of reduced-resolution subimages derived from said encoded original image; means for structuring the encoded original image and the encoded reduced-resolution subimages within said image-pyramid data structure as a 49 o: 15 plurality of individually-accessible tiles; o* means for accepting an image-resolution specification entered by a user, the image-resolution specification defining an image resolution within a range of image resolutions spanned by the image-pyramid data structure; and means for selecting, accessing, and interpolating between tiles respectively from two adjacent levels of the image-pyramid data structure to generate data encoding the first image portion at an image resolution corresponding to the image resolution defined by the image-resolution specification; means for capturing image-editing commands entering by a user for editing 090e* 0 25 said first image portion by modifying the first image portion corresponding to imposing the image-editing effects specified by the image-editing commands on the first image portion, each image-editing command being representative of one of a plurality of editing modes; means for generating editing-effect parameters that define the modifications to said first image portion corresponding to imposing the image-editing effects specified P \OPER\ AA64038.94.147 1216/9 -8by the image-editing commands.

means for storing said editing-effect parameters in an editing-effect data structure representative of a plurality of layers, each layer being associated with an editing mode corresponding to the editing mode associated with the editing-effect parameters stored in the layer; and means for generating data encoding a second image portion by combining said editing-effect parameters in the editing-effect data structure with data encoding said first image portion to impose the image-editing effects defined by said parameters on the first image portion so that in operation said second image portion may be displayed on the display means to display the first image portion as modified by image-editing effects specified by the image-editing commands.

Embodiments of the invention provide a particularly efficient Ptethod of image processing. It also offers unexpected possibilities as to the scale of the elementary operations that can be carried out on an image, all this with an extremely reduced processing requirement.

Embodiments of the invention enables the operator to follow the changes made to the image as they are carried out, almost instantaneously. In addition, the operator can, at any moment, return and redo an elementary operation, again almost instantaneously. This is because the different actions and their results (that is, the layers) are defined by very simple equations, taking up little memory space, and the function managing the synthesis of the operations carried out up to that moment is again defined by an equation with an extremely simple structure.

In this way, embodiments of the invention provides for any kind of image effect, such as airbrushing, accentuations of definition or contour, dissolve effects, color modifications, in short any operation concerning image graphics and color. Embodiments of the invention also enable geometrical transformations or modifications, such as rotation, changes of scale, etc. All these can be followed in real time by the operator, using in general a mouse or light pen on an interactive tracing table.

This pen may provide two types of command signal: one is a position signal giving the coordinates of the dot concerned, and if necessary its environment (for example the P \OPER\KAT64038-94.147- 1216W7 -9path of an airbrush stroke); the other uses the pressure of the pen on the table to create a second type of signal (in.the airbrush example, it would govern the density of the color being "sprayed").

Each elementary operation may be identified separately and be retrieved from memory at any moment. The sum of all these elementary operations may be constantly updated as the work evolves. In this way the operator can access, at any moment, either the overall result of all the operations carried out up to that moment, or the intermediate operations identified in the layers. The operator may therefore intervene and modify a layer without in any way affecting the other layers. The link between the layers may be only at the level of the recurrence.

When all the necessary operations are finished, and the operator wishes to produce the final image or an intermediate one at a given definition, he or she may order a functional interpolation at the required image definition. When the fimnction has been defined in this way at all the dots at a given definition, the system may calculate the pixels at these different points ,,ing the functions obtained by interpolation.

The number of dots (nodes to which it is necessary to define the elementary functions) and the size of global function may, in general, be relatively small because the function evolves with little variation (its second derivative is generally very low for most of the dots in the image). The function only varies substantially at dots corresponding to a large color change.

The grid chosen for the definition of elementary functions can be one with a equal mesh at all points. It can also be constructed using a different sized mesh at various points, depending on whether the image zone covers an area of small or great variation. This makes processing and correction easier.

Even if the final image is unsatisfactory, once the control run has been carried out and (for example) a proof image has been printed, it is still possible to go back and correct any of the intermediate stages in order to achieve a better result.

The invention is described below in a more detailed way, using examples.

In general, the invention provides processing of an image, that is the data making up the image. The data may be obtained by scanning film, printed images, etc...

I

P \OPERA IM038,94.147. 126/97 The definition of such a scan is selected in relation to the final image to be obtained, but as the process is usually intended to produce films for monochrome or color printing, a high level of definition is required.

This large collection of data is then sampled at a definition matching that of the monitor screen being used by the operator to apply modifications to the image. Such screens always have a very low definition in relation to that of either the original image or the image required in output. The sampling of the original image is carried out in relation to a grid, whose ends or nodes match the coordinates of the pixels displayed on the screen. Only the original image data matching these nodes will be used for the elementary or intermediate processing operations.

The result of each elementary operation i is known as a layer i. The intermediate result corresponding to all the operations carried out, including the final one, up to the layer i can be displayed for checking on the screen.

In fact, according to the invention, the result of an intermediate operation in progress appears as it is executed, in real time.

Only when all the elementary operations have been carried out to the operator's satisfaction will the operator decide to create the final image. To do this he or she will use the data of the original image in order to provide the definition :.',uired for the final image.

The result of an elementary operation on a displayed or edited image is in fact the result of the sum of all the previous elementary operations.

The functions representing the elementary operations are added to arrive at a single global function in the same form as the previous elementary operations. As the functions of the elementary operations are only defined at a limited number of dots (that is, at the selected nodes) the combination of the functions takes place only at the nodes.

To edit the image, either for display on the monitor or to produce a final proof, according to the invention, a calculation is made of the value of the global function at each dot at which it is necessary to create a pixel, depending on the definition required for the edition. This is done by function interpolation.

By convention, in this description, we shall talk about the final image, even though $0O3 this final image will usually be made up of three images corresponding to the three primary P:\OPER\KAT6038 94 147 12/6/97 11 colors (in the case of color printing) with an additional image in some cases for the grey tones.

According to the invention, the processing to be carried out on the image covers operations which vary greatly in terms of both color and geometry. These include airbrushing, modification of tones or colors (or changes to these), or geometric changes relating to all or part of the image. This can also include importing images from other sources to make a kind of collage. An image built up in this way may also, in turn, be modified in its entirety.

The processing can also be applied to the geometry of all or part of the image. In this way the operator can deform parts of the image to give the impression of movement, anamorphosis, etc...

Operations may also cover the creation of backgrounds or superimpositions such as skies or clouds, water, river or wave effects, smoke or mist effects, snow, rain, sun, day, night; blurred or highly contrasted images, or multiple combinations of all or any of these.

Each elementary operation corresponds to an image "layer", similar to oil painting technique.

This elementary operation is defined by the following general equation associated with the layer i.

**This general function is as follows; (1) t* p(xy) a,(xy) (xy) P,(xy) y,(xy) This function applies to a dot or pixel with coordinates x,y in the reference plan. In fact, as described above, the variables x and y vary in a discontinuous manner because the processing relates only to the grid nodes. The function pi is the transformation number i which will be carried out at the dot with the coordinates x,y.

This general function is recurrent because it takes into account the existing image in the layer i-i. This image is defined by the function pi.i This function also provides for the importation of an external image Ii. This import function is defined by a scaler ,3 indicating the total or partial presence or absence of P.\OPER\KATV64038.94.147- 121/6 -12a dot of the imported image I i at the location The import function Pi defines the import operation. Usually this is a geometrical operation carried out on the dots of the image to be imported I1 to place it in the layer i.

This import function Pi can be simple translation, a rotation or a combination of operations, including anamorphic deformation.

The final term of the function y, is the chromatic modification of the dot (x,y) in the layer i.

In this way, the function cp defines the transformation of the dot with the coordinates depending on the elementary operation to be carried out. This transformation modifies the previous state of the dot (layer i-1) obtained by the function p.1 to add to this dot the components of the dot of the imported image I and a possible chromatic modification by the term y,.

Depending on the elementary operations carried out, the value of the scalars of the two first terms will be different, or in certain cases complementary. The third term may be nil in some cases.

To simplify, we shall consider that the first layer is a blank sheet with no image and with a uniform color yO.

According to the invention, we consider that the image to be processed is an image to be imported, also that the second layer may consist in importing the original image ill: be If the following operations are limited to the processing of this original image, there will be no other image to be imported. If on the other hand this original image has to be supplemented by overlaying images or parts of images, rather like a collage, there will be other imported images I,.

In order to display the intermediate result, that is, the transformation of the image arising from the elementary operations, that is, for all the elementary layers from 1 to i, it is necessary to add the layers i and obtain a general equation which matches the synthesis of the layers. This general equation is as follows: q number of images imported.

I' IR\KAT\64038.94.147- 12/6/9 -13- Esu I Y(xy) (3) It will be noted that this general equation has an extremely simple form, equal to the sum of the terms relative to the images imported and the chromatic coefficient. In this way, for three imported images, the general function will be limited to four terms. Such a simple general expression takes up only a very small part in memory and the calculation of all the dots of the final image can also be done in a very simple way by calculating the parameters of the general equation at the different dots necessary to the final or intermediate image.

When the intermediate image has to be displayed on the control screen (an automatic procedure, necessary for the operator), the intermediate image, the result of the elementary operation, appears almost immediately on the screen, as the operation is carried out. This is because the number of nodes on the grid is very limited.

For the final image (for example intended for print film), the calculation time is longer (as the definition is a thousand times higher), without being prohibitive. The operator can :check the final result at the desired definition. If this is unsatisfactory, the operator can review the layers i and modify this or that elementary operation, accessing elementary 15 operations immediately (as these are defined by a function retained in memory).

S" When the operator has corrected or modified the appropriate layer, he or she can, again, easily create a synthesis of the layers to obtain the final image, verifying on the proof the result of the correction. This is preferable to checking on the monitor only.

:The following description will be limited to a few elementary operations, given that 20 the possibilities are virtually endless even though the form of an elementary operation is limited to a very simple equation.

First of all, in order to set up the grid, instead of choosing a grid with the same pixel definition as that of the monitor screen, a grid is selected with a definition four times smaller.

The only dots selected will be the four corners of a quadrilateral each of whose sides corresponds to four pixels. These comers are known as nodes while the whole quadrilateral is known as a mesh.

1) Airbrushing: P:OPER\KAT\6038.94.147 -12/6/91 -14- This consists in making a line with a color. As this line imitates that made by an airbrush it can be treated as a succession of colored dots created by the airbrush spray. The distribution of the color density in a airbrush dot is a Gauss function. This means that the intensity of the color is at its greatest in the center of the dot, diminishing towards the edges as a Gauss function. In a real airbrush, the intensity depends on the pressure exerted on the trigger, which widens or otherwise the ink spray within the airjet. Such a pressure can be simulated in a computerised system by representing (as explained above) a dot by a circle of color with a density variation between the center and edge expressed as a Gauss function.

The saturation at the center can vary between 0 and 1 (or 0 and 100%).

To sum up, the line of an aerograph is a succession of colored disks, of which it is possible to modify the path (the location of the disk centers), and the color density.

Based on the general equation and the airbrush characteristics, P(xy) a,(xP,- 1 (x Y,(xY) P,(xy) OV(xy) Y,(xy) C SC color constant of the "projected material" this equation becomes the following: p,(xv) ai(xy)(p-I(xy) C As there is no imported image in the path of the airbrush, the coefficient of presence 3i of an external image is nil at all points of the layer.

The application of the airbrush consists in replacing partially or totally the previous shade of a dot by the shade of color "projected" by the airspray. Because of this, the chromatic function y, is expressed as a function of the color C and as a complement 1 to the coefficient of presence of the previous image, that is cx 1 I a 1 P \OPER\KATO38M. 147 12//97 The choice of scalar ai at each dot translates the density of color left by the airbrush.

The function of color presence ai can be represented by a Gauss function centered on one dot, limited for example to 10% at the edge of the disk. In other words, the two extreme ends of the Gauss curve beyond 10% (or any other value which may be selected) are suppressed. This means that the Gauss function will not be applied beyond the disk radius chosen.

2) Image fusion: This operation consists in importing an external image into an existing one. Based on the general equation, this import operation is defined as follows: In the general equation to which are applied the particular conditions relating to this operation: P,(xy) IPi(xy) 0 3,(xy) a The chromatic function Y, is nil and the coefficients a, and P are 15 complementary coefficients (their sum is equal to one).

In fact, as a hypothesis for this type of operation, a dot of the imported image replaces, more or less, or even completely, a dot of the previous image. This corresponds in the first instance to a more or less pronounced dissolve and in the second to the replacement of the part of the previous image within the contour of the imported one.

The equation below can be simplified and thus gives the equation for image fusion: a (xy) i,(xy) IP,(xy) 3) Lightening/darkening It should be noted that in the general equation of a layer i, the scalar al is never nil at all points of the layer. On the other hand, if there is no image importation, the scalar Pi is P \OPER\KAi'\ 038-94.147 12/6/ -16nil at every point To lighten or darken an image, it is necessary to use the chromatic function y; As explained above, the general function (pi is never limited to the chromatic function, for this would mean suppressing all the images in layers 1 to i-1 (disappearance of that is, the recurrence.

The darken/lighten function therefore consists in adding a color to the color at the previous dot x,y (function of Based on the general equation, as follows: p,(xy) a (xy) p 1 P,(xy) y,(xy) in which a,(xy) 1 V(xY) P,(xy) 0 V(xy) o S We obtain:

S

y,(xY) 4) Deformation/anamorphosis: This operation is applied to an imported image. In fact, if it is desired to transform 15 part of the image of the layer this part of the image is considered as an imported image to be treated as described below.

The deformation/anamorphosis of an image consists of linking to each node a vector of deformation with a direction and size corresponding to the desired deformation. If this 0oo deformation is uniform over all the relevant part of the image, this will mean that each node will have attached to it vectors of the same size and direction. These will move the dot corresponding to each node as defined by each vector.

To achieve such a deformation, the general function of the layer i becomes as follows through the use of the equation defining image import: P \OPER\KAT\6038-94.147 12/6/J7 -17- (Pi(XY) a(xY) (P-l(Xy) i 1 (xy) IP,(xy) The deformation or anamorphosis consists in working on the import function P, Levelling: This consists in levelling a color in part of an image. As an example, in a portrait, it would enable tie operator to remove local skin defects, such as birthmarks. To achieve this, it is necessary to calculate the average intensity of the color in a disk centered on each node of the part of the image to be processed. Depending on the radius selected, the color will be made more or less uniform.

In fact, this operation consists in combining the normal image with another which has been averaged out.

6) Contrasting: This is the opposite to the previous type of processing. It involves accentuating the fineness of the lines in a drawing or photograph. In a portrait, for example, it would bring out individual hairs of a hairstyle.

To achieve this, it is necessary to increase the high-frequency wavelength harmonics 15 without touching the low frequency ones (near the average).

t* 0* «o• I'-P I

Claims (3)

1. A method of processing a digital image in a computerised system in order to obtain an image to be used for display or printing, in such a way that the checking of each elementary operation is effected by displaying the result of this operation on a monitor screen, the procedure being characterised by the following: the original image to be processed is sampled in accordance with a definition grid so as to retain from all the dots contained within the original image only a number of dots equal at the most to the number that can be displayed on the monitor screen, the processing of the image is broken down into elementary recurrent operations each broken down in turn into three parts and providing, based on the result of the previous elementary operation (layer a new result (layer these three parts added to each other representing: adopting the new layer of a color dot from the previous layer (i-1) with a weighting ranging from 0 to 100% (positive or negative value), importing an external image (Ii) into the layer i, by importing of a color dot from the image after chromatic and geometric transformation of this dot S. to add it to the color dot of the layer the degree of replacement of the dot of the layer by the dot imported from the image (Ii) being defined by a scalar (Pi(x,y)) with values from 0 to 100% (positive or negative values), carrying out a chromatic modification on the dot of the layer .each elementary operation being defined by the equation taking account of the previous operation *e a e a(xy) P3,(xy) y,(xy) in which: Si(x,y) is a scalar function of the dot corresponding to the presence at this dot t4 Li2 P'\OPERKAT'\1038.4. 147 12/6/9

19- of the image resulting from the previous elementary operation (pi(x,y) is a function representing the previous elementary operation, Pi(x,y) is a scalar function corresponding to the presence at dot of a dot corresponding to the imported image, Ii represents the imported image made up of a set of dots, Pi(x,y) is the function of image import representing the chromatic geometric transfer of one of the set of dots in the image towards the layer to which is applied the elementary operation pt(x,y), Ii[Pi(x,y)] is the function corresponding to the import of the image, i(x,y) is a chromatic function representing a color transformation function carried out on a dot 4* each of the terms pJ 1 i[Pi(x,y)] and Yi(x,y) not being precluded from being nil while the term ai(x,y) cpi-(x,y) never being nil for all the dots S" 2. A method as claimed in claim 1 wherein elementary operations are effected to obtain a function representing the i first elementary operations in order to obtain a function whose parameters are defined at all the dots of the definition grid 4 J=q E I y(xy) j=1 q number of imported images, in this global function: LI P \OP!IR\IKA'64038-94 147 12/6/97 aj(x,y) is a scalar analogous to the scalar ai(x,y) of a elementary function Ij represents an image j to import Pj(x,y) is an import function analogous to the previous import functions Pi(x,y) y(x,y) is a chromatic function analogous to chromatic functions the global function being defined by interpolating it at the intermediate dots between the dots of the definition grid, these intermediate dots depending on the definition required for the final image, the pixels being calculated for each dot to be obtained. 3. Apparatus for editing, in a digital image processing system, an original image comprising: display means for displaying image portions; means for generating a first image portion at a user-specified resolution, including: 1) means for generating and storing an image-pyramid data structure *encoding the original image and a plurality of reduced-resolution subimages derived from said encoded original image; i means for structuring the encoded original image and the encoded reduced-resolution subimages within said image-pyramid data structure as a plurality of individually-accessible tiles; means for accepting an image-resolution specification entered by a user, the image-resolution specification defining an image resolution within a range of image resolutions spanned by the image-pyramid data structure; and means for selecting and accessing one or more tiles of said encoded original image or reduced-resolution subimages to generate data encoding the first image portion at an image resolution corresponding to the image resolution defined by the image-resolution specification; P \OPER\KAT\64038-I4.147 12169 -21- means for capturing image-editing commands entered by a user for editing said first image portion by modifying the first image portion corresponding to imposing thdie image-editing effects specified by the image-editing commands on the first image portion, each image-editing command being representative of one of a plurality of editing modes; means for generating editing-effect parameters that define the modifications to said first image portion corresponding to imposing the image-editing effects specified by the image-editing commands; means for storing said editing-effect parameters in an editing-effect data structure representative of a plurality of layers, each layer being associated with an editing mode corresponding to the editing mode associated with the editing-effect parameters stored in the layer; and means for generating data encoding a second image portion by combining said editing-effect parameters in the editing-effect data structure with data encoding said first image portion to impose the image-editing effects defined by said parameters on o the first image portion so that in operation said second image portion may be :displayed on the display means to display the first image portion as modified by image-editing effects specified by the image-editing commands. 4. The apparatus of claim 3, wherein said second image contains image information from said selected tiles. 5. The apparatus of claim 3 further comprises means for compressing image information within at least one of said tiles in said reduced resolution subimages or said original image. a 6. A method for editing an original image, in a digital imaging processing system, comprising the steps of: generating and displaying a first image portion at a user-specified resolution, including the steps of: 1) generating and storing an image pyramid data structure encoding the -I- P \OPEI\KA'T6403894 147 12/6/9W -22- original image and a plurality of reduced-resolution subimages derived from said encoded original image; structuring the encoded original image and the encoded reduced- resolution subimages within said image-pyramid data structure as a plurality of individually-accessible tiles; accepting an image-resolution specification entered by a user, the image-resolution specification defining an image resolution within a range of image resolutions spanned by the image-pyramid data structure; selecting and accessing one or more tiles of said encoded original image or reduced-resolution subimages to generate data encoding the first image portion at an image resolution corresponding to the image resolution defined by the image-resolution specification; and 5) displaying the first image portion so generated; capturing image-editing commands entered by a user for editing said first image portion by modifying the first image portion corresponding to imposing the image-editing effects specified by the image-editing commands on the first image :portion, each image-editing command being representative of one of a plurality of editing modes; generating editing-effect parameters that define the modifications to said first image portion corresponding to imposing the image-editing effects specified by the iQ image-editing commands; storing said editing-effect parameters in an editing-effect data structure representative of a plurality of layers, each layer being associated with an editing mode corresponding to the editing mode associated with the editing-effect parameters stored in the layer; generating data encoding a second image portion by combining said editing- effect parameters in the editing-effect data structure with data encoding said first image portion to impose the image-editing effects defined by said parameters on the first image portion; and displaying said second image portion to display the first image portion as P I' )l'il\KAT\(4O3894 1417 12/1( -23- modified by image-editing effects specified by the image-editing commands. 7. The method of claim 6 wherein said second image contains image information from said selected tiles. 8. The method of claim 6 further comprising the step of compressing image information from said tiles within said reduced resolution subimages or said original image. 9. Apparatus for editing an original image in a digital image processing system comprising: display means for displaying image portions; means for generating a first image portion at a user-specified resolution, including: 1) means for generating and storing an image-pyramid data structure encoding the original image and a plurality of reduced-resolution subimages derived from said encoded original image; means for structuring the encoded original image and the encoded reduced-resolution subimages within said image-pyramid data structure as a plurality of individually-accessible tiles; means for accepting an image-resolution specification entered by a i user, the image-resolution specification defining an image resolution within a range of image resolutions spanned by the image-pyramid data structure; and means for selecting, accessing, and interpolating between tiles respectively from two adjacent levels of the image-pyramid data structure to generate data encoding the first image portion at an image resolution corresponding to the image resolution defined by the image-resolution specification; means for capturing image-editing commands entering by a user for editing said first image portion by modifying the first image portion corresponding to imposing the image-editing effects specified by the image-editing commands on the I- P\OPER\KAT\61038-94.147. 1216/9 -24- first image portion, each image-editing command being representative of one of a plurality of editing modes; means for generating editing-effect parameters that define the modifications to said first image portion corresponding to imposing the image-editing effects specified by the image-editing commands. means for storing said editing-effect parameters in an editing-effect data structure representative of a plurality of layers, each layer being associated with an editing mode corresponding to the editing mode associated with the editing-effect parameters stored in the layer; and means for generating data encoding a second image portion by combining said editing-effect parameters in the editing-effect data structure with data encoding said first image portion to impose the image-editing effects defined by said parameters on the first image portion so that in operation said second image portion may be displayed on the display means to display the first image portion as modified by image-editing effects specified by the image-editing commands. 10. A method for editing an image substantially as hereinbefore described with reference to the accompanying drawings. 11. Apparatus for editing an image substantially as hereinbefore described with reference to the accompanying drawings. DA TED this 12th day of June, 1997 BRUNO DELEAN By his Patent Attorneys Davies Collison Cave C C C By his Patent Attorneys INTERNATIONAL SEARCH REPORT intern lApplicaio No PCT/FR 94/00313 A. CLASSIFICATION OF SUBJECT MATTER IPC 5 G06F15/72 According to International Patent Classificaton (IPC) or to both national classificaton and IPC I B. FIELDS SEARCHED i Minimum documentation searched (classfication system followed by classfication symbols) IPC 5 GO6F Documentation searched other than minimum documentation to the extent that such documents are included in the fields searched Electronic data base consulted during the international search (name of data base and, where practical, search terms used) C. DOCUMENTS CONSIDERED TO BE RELEVANT Category' Citation of document, with indication, where appropriate, of the relevant passages Relevant to claim No. Y GERARD J. HOLZMANN 'beyond photography.the 1 digital darkroom' 1988 PRENTICE HALL SOFTWARE SERIES USA see page 27 page 29; figures 3.29-3.33 Y EP,A,O 198 269 (DAINIPPON SCREEN) 22 1 October 1986 see page 3, column 4, line 10 line 22 see claim 1 A WO,A,92 18938 (SCITEX AMERICA CORP.) 29 1 October 1992 see page 21, line 25 page 26, line figures 2-3C A EP,A,O 462 788 (LINK-MILES) 27 December 1991 /i- j Further documents are listed in the continuation of box C. Patent family members are listed in annex. *Special categories of cited documents: 'T later document published after the international filing date or priority date and not in conflict with the application but document defining the general state of the art which is not cited to understand the principle or theory underlying the considered to be of particular relevance invention earlier document but published on or after the international document of particular relevance; the claimed invention filing date cannot be considered novel or cannot be considered to 'L document which may throw doubts on priority claim(s) or involve an inventive step when the document is taken alone which is cited to establish the publication date of another document of particular relevance; the claimed invention citation or other special reason (as specified) cannot be considered to involve an inventive step when the document referring to an oral disclosure, use, exhibition or document is combined with one or more other such docu- other means ments, such combination being obvious to a person skilled document published prior to the international filing date but in the art. later than the priority date claimed document member of the same patent family Date of the actual completion of the international search Date of mailing of the international search report 4 July 1994 05. 07, 9 Name and mailing address of the ISA Authorized officer European Patent Office, P.B. 5818 Patentlaan 2 NL 2280 HV Rijswijk Tel. (+31-70) 340-2040, Tx.31 651 eponl Perez Molina, E Fax: 31-70) 340-3016 Form PCT/ISA/210 (econd sheet) (July 199) page 1 of 2 INTERNATIONAL SEARCH REPORT IItr lApiainN PCT/FR 94/00313 C.(Continuation) DOCUMENTS CONSIDERED TO BE RELEVANT Category 'Citation of document, wih indication, where appropriate, of the relevant passages Relevant to claim No. A EP,A,0 512 839 (QUANTEL) 11 November 1992 2 Form PCT/ISA/210 (oantinusiton of second sheet) (July 1992) page 2 of 2 INTERNATIONAL SEARCH REPORT ApchnN *..irmahnonatenniiym~mersPCT/FR 94/00313 Patent document I Pblicaio Patent family Publication cited in search report -P date member(s) date EP-A-0198269 22-10-86 JP-B- 4076263 03-12-92 JP-A- 61227477 09- 10-86 JP-B- 6018433 09-03-94 JP-A- 61227480 09-10-86 EP-A- 0359292 21-03-90 US-A- 4712141 08-12-87 WO-A-9218938 29-10-92 EP-A- 0543964 02-06-93 ~JP-T- 5508507 25-11-93 EP-A-0462788 27-12-91 GB-A- 2245460 02-01-92 EP-A-0512839 11-11-92 GB-A- 2256557 09-12-92 JP-A- 5161065 25-06-93 F~orm PCT/ISA/210 (patent family annex) (July 1992) RAPPORT DE RECHERCHE INTERNATIONALE CLASSEMENT DE LUOBJET DE LA DEMANDE CIB 5 G06F15/72 Dcm, Intcmationale No PCT/FR 94/00313 Scion la classification intemnationale des brevets (CIII) ou A la fois selon la classification nationale et la CIII B. DOMAINES SUR LESQUELS LA REC14ERCHE A POwR Documentation miunimale consultbe (syst~me de classification suivi dcs syinboics de classemencrt) CIB 5 G06F Documentation consultec autre quc la documentation rninsmalc dans la rncsue oQ ces documents rel~vcnt des domaines stir lesqucis a port6 la recherche Base de donrits Clectronique consultte au courm dc Is recherche internationale (nom dec la base de dorntes, ct si cela est rtaiisable, termes de recherche utiuasts) C. DOCUMENTS CONI;4)3RES COMME, PERTINENTS Categone' Identification des documents citts, avec, le cas tchtant, I'mndication des passages pertinents no. des rcvcndications vistes Y GERARD J. HOLZMANN 'beyond photography.the1 digital darkroom' 1988 PRENTICE HALL SOFTWARE SERIES USA voir page 27 page 29; figures 3.29-3.33 Y EP,A,O 198 269 (DAINIPPON SCREEN) 22 1 Octobre 1986 voir page 3, colonne 4, ligne 10 ligne 22 voir revendication 1 A WO,A,92 18938 (SCITEX AMERICA CORP.) 29 1 Octobre 1992 voir page 21, ligne 25 page 26, ligne figures 2-3C A EP,A,0 462 788 (LINK-MILES) 27 Ddcembre 1991 MV Voir la sute du cadre C pour la fin de la liste des documents Les documents de families de brevets sont mndaques en annexe *Cattgories speciales de documents cit 'T document ultkrseur publit aprts la date do international oti la A' dcumnt dlimsantl~tt geera de a tchniuenondate dc priontt et n'appartenenant ps A ltatde la W document cominiesn p ta tcu emet~ perltecntqe o technique pertinent, mass cite pour comprendre le pnincipe consdtr come prtiulimmen pctin-ntot la thsoone constituant la base de l'mnvention W document anttneur, mass publi6 A la date de dtp~t international X dcmn atcl~eetprielisvninrvniuen ei ou airs cett date tre considtrc comrnc nouvelie oti cornme impiiquant tine activitt document pouvant jeter un doute stir une revendication de inventive par rapport aui document consd~re isoitment priontt ou citt pour d~trminer Is date de publication d'usse document particulierement pertinent; I'invention revendiquce sute citation oti pour sine raison specisie (tele qu'indiqu~) ne peut ate consideree comme inipliquant tine actite inventive docuiment se r~ftrant Autne divulgation orale, Autnusage, A lorsque le document cit associ6 Ai tn oti piusietirsasuites; tine exposition oti tous autres moycns documents de meme nature, etie combinaison 6tant tvidente Pdocumnent pubiC avant la date de dep~t international, mass pour tine personne du mftier poistricurernent A la date de pnont6 revendiqu~c W. documnent qua fait partic de la marne fasnaile de brevets Date A laquelle la recherchse intemnationale a ett effeclivemnrt achcvtc Date d'exptidition du prtsent rapport de recherche interaionale 4 Juillet 1994 05- 07,94 Nomn et adresse postale de I'administrstion charg~tic dea recherche intemationale Fonctionnaire autorist Office Europten des Brevets, P.B. 5818 Patentlaan 2 NL 2280 HV Riisnik Tel. +31-70) 340-2040, Tx. 31 651 cpoill,Pe z Moia E I

31-70) 340-3016Pe zMo ia E Ponisuizire PCT/ISA/2ii (deuxiimse feuille) (Isiucet 1992) page 1 de 2 RAPPORT DE R13CHERCHE INTERNATIONALE Dem Itirnamsnic No PCT/FR 94/00313 C.(suste) DOCUMENTS CONSIDERES COMME PERTINENTS Cattgone 'Identification des documents citte, avec, It cas 4tchtait, 'ijndication des passages pertinents no. des r-evcndication vistes A EP,A,0 512 839 (QUANTEL) 11 Novembre 1992 Forenutali PCtISA,'210 (sUt de la deuxiimm feuillt) (Cillet 1992) page 2 de 2 RAPPORT DE RECH-ERCH-E INTERNATIONALE Dema -atcMaronale No Renscignements relaofs aux n,.mbres de farnzlles de brevets I PCT/FR 94/003 13 Document brevet citi Date de Membre(s) de la Date de au rapport de recherche T publication famille de brevet(s) publication EP-A-0198269 22-10-86 JP-B- 4076263 03-12-92 JP-A- 61227477 09-10-86 JP-B- 6018433 09-03-94 JP-A- 61227480 09-10-86 EP-A- 0359292 21-03-90 US-A- 4712141 08-12-87 WO-A-9218938 29-10-92 EP-A- 0543964 02-06-93 JP-T- 5508507 25-11-93 EP-A-0462788 27-12-91 GB-A- 2245460 02-01-92 EP-A-0512839 11-11-92 GB-A- 2256557 09-12-92 JP-A- 5161065 25-06-93 Formulaire PCT/ISA.1210 (annexe famlUes de brawts) (uillet 1992)