Facsimile and static presentation processing – Static presentation processing – Attribute control
Reexamination Certificate
1999-11-29
2004-02-03
Lee, Thomas D (Department: 2624)
Facsimile and static presentation processing
Static presentation processing
Attribute control
C358S003130, C358S003200
Reexamination Certificate
active
06687021
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention is in the field of electronic reproduction technology and is directed to a method for the production of rastered color separations with rasters having arbitrary screen widths and screen angles.
Tonal gradations cannot be undertaken in most printing processes because ink is either applied onto the printed matter or not in these processes. Color dots arranged in grid-like fashion are therefore printed next to one another for simulation of tonal gradations. A brighter or darker tone optically arises by varying the size of the color dots. These color dots are also referred to as half-tone dots, their spacing from one another as representing screen width and the angle of the half-tone dot grid relative to the x-axis representing screen angle. Work is usually carried out with four rasters having the same screen width and the screen angles of 0, 15, 45 and 75 degrees for minimizing what is referred to as moiré for four-color printing with the inks cyan, magenta, yellow and black. Even minimal deviations therefrom cause moiré and, thus, poor or unusable image reproductions.
The rastered print masters are produced with digitally working exposers in the Prior Art. These use a light source in order to illuminate a light-sensitive surface. This light source can be displaced in steps in the x-direction and the y-direction relative to the print master and can be designationally turned on and off. Small surface elements, also called device pixels, thereby derive, these being capable of being individually illuminated or not dependent on the control of the light source. Half-tone dots are constructed of a plurality of device pixels given these digital exposers. Each device pixel is represented by a bit in a memory in a preceding computer unit, also called a RIP (raster image processor). The totality of these bits is referred to as a bit map and is a digital image of the device pixels.
A number of methods for digital rastering have become known in the past, these being capable of being divided into 3 groups, what are referred to as irrational rastering methods (referred to below as IS, IS method or IS technique), what are referred to as rational rastering methods (referred to below as RT, RT method or RT technique) and frequency-modulated rastering methods. These names can be explained based on the fact that screen angles having an irrational tangent are employed in the irrational rastering methods and screen angles having a rational tangent are employed in the rational rastering methods. For example, an irrational rastering method is disclosed by German Letters Patent DE 28 27 596 C2, and rational rastering methods are disclosed by German Letters Patent DE 28 27 596 and by European Patent Publication Number 0 539 387 B1.
What these two rastering methods have in common is that a digital image of the raster dots exists in the memory of the RIP, this also being called “threshold hill” or threshold matrix. The term “threshold memory” is also used in this context. This image is composed of a group of numerical values, what are referred to as the thresholds. For producing the digital print masters, also referred to as a rastering process or a rastering, a threshold is selected from the threshold hill for each bit of the bit map according to a rule dependent on the rastering method and is compared to a digital color tone signal allocated to the bit of the bit map, which is usually acquired by scanning an original or is read out from a memory, and the corresponding bit is set or not dependent on this comparison and, thus, the corresponding device pixel of the exposer is also blackened or not. The procedure of defining the threshold allocated to a bit is also called threshold access.
In the IS method, the threshold memory is composed of the digital image of a single raster dot that is deposited under an angle of 0 degrees. In the rastering, a complex calculating rule must be implemented for each bit to be rastered. Both the screen width as well as the screen angle—also referred to only as raster below—must be taken into consideration.
In the RT method, the threshold memory is composed of the digital image of one or more raster dots. Screen width and screen angle are already taken into consideration when the thresholds are deposited in the memory, and the thresholds are organized in the memory such that the thresholds of neighboring bits are also neighbors in the threshold memory. The calculating process given a threshold access is thus essentially reduced to an address incrementation.
The two methods have a completely different technical realization as relates to the method of execution for producing the rasters. On the other hand, the two methods have different technical advantages and disadvantages, for which reason the one rastering method is sometimes favored in practice, and sometimes the other. Since only a single raster dot is deposited in the threshold memory given the IS method, there is a far lower need for memory space. Further, the calculating process for the selection of the thresholds allows the exact realization of all rasters, particularly the irrational angles 15 and 75 degrees that are important for color printing. The RT methods, by contrast, can only approximate the demanded screen angles with values that can be described by a rational tangent, whereby the need for threshold memory space increases approximately quadratically relative to the precision. The advantage of the RT method is comprised therein that the rastering process can be realized with less expense both in terms of hardware and software and better performance can be realized, whereby the IS methods fare especially badly in the realization in software. This is especially true when large areas are to be rastered with a uniform degree of surface coverage. For the RT methods, a bit pattern that corresponds to the degree of surface coverage can be produced, and this bit pattern then only has to be copied. For the IS method, each bit to be rastered, by contrast, must fundamentally be separately calculated. In the RT methods, moreover, it is not necessary to add noise. Noise is usually added in the IS methods in order to cover artifacts that arise given these methods. In practice, this leads to raster dots with torn edges that have a disadvantageous effect in terms of printing technology. Over and above this, the noise in large areas with constant surface coverage is expressed in the form of an optical restlessness of the reproduced image.
U.S. Pat. No. 5,315,407 discloses an improvement of the IS method with which the restlessness that is produced due to the somewhat different shape of neighboring raster dots is to be reduced. To that end, the coordinates for addressing the threshold matrix are modified such that the midpoint of every raster dot is shifted onto the respectively closest device pixel. As a result thereof, the raster dots are lent a more uniform shape. This method, however, also exhibits the fundamental disadvantage of the IS method that a complex calculation of the addresses for the threshold access is to be calculated for each bit to be rastered.
SUMMARY OF THE INVENTION
An object of the invention is to specify an improved rastering method that avoids the disadvantages of the known methods and assures a high reproduction quality.
The invention achieves this in that a raster like that disclosed by European Letters Patent EP 0 539 397 B1 is approximated by a threshold hill, and that a continuous error calculation during the rastering process describes the deviation from the required rated value and, given upward transgression of an error tolerance limit, continues the rastering at a different location of the threshold hill, so that the error sum that has thereby accumulated is compensated.
The method disclosed here unites the advantages of the two rastering methods with one another in an optimum way. As in the IS technique, on the one hand, the method allows work to be carried out with slight threshold memory space and to exactly reproduce arbitrary rasters, and, as in
Heidelberger Druckmaschinen AG
Lee Thomas D
Schiff & Hardin & Waite
LandOfFree
Method for digitally rasterizing half-tone images with... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for digitally rasterizing half-tone images with..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for digitally rasterizing half-tone images with... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3294652