Reexamination Certificate
1998-11-10
2002-05-28
Grant, II, Jerome (Department: 2624)
C382S169000
Reexamination Certificate
active
06394612
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to image process apparatus and method for quantizing multivalue original image information into a recording level of a recording device.
2. Related Background Art
Recently, various office automation (OA) equipments such as a personal computer, a word processor and the like have come into wide use. In this connection, various recording methods such as an ink jet recording method, an electrophotographic recording method, a wire dot recording method and the like have been developed as methods to print out information input by the OA equipments. In such the current recording methods, a so-called binary recording method to record or not record a dot (i.e., recording pixel) on a recording medium is the mainstream. However, according to a recent advance in functions of the personal computer, the word processor and the like, it has popularized to output a photographic image, a desktop publishing (DTP) image or the like as a recording image. For this reason, it has been strongly expected to realize pure and fluent halftone representation.
A dither method and an error diffusion (ED) method have been known as typical methods to represent a halftone in such a binary recording device. Therefore, these two methods will be briefly explained as follows.
The dither method is a gradation representation method to represent gradation or tonality by initially determining a unit matrix for representing the gradation and then controlling the number of recording pixels to be recorded within the unit matrix. For example, in case of defining a 4×4 unit matrix, the number of recording pixels to be recorded within the defined matrix is controlled to be “0 dot” to “16 dots”, thereby realizing 17-level gradation representation.
On the other hand, a principle of the ED method has been published in R. W. Floyd and L. Steinberg, “An Adaptive Algorithm for Spatial Gray Scale,” SID 75 Digest (1976). That is, the ED method is the gradation representation method to perform quantization by calculating a density difference (i.e., error data) between an original image pixel density of each original image and a recording pixel density to be recorded by the recording device after the quantization, and diffusing the calculated error data to peripheral pixels before the quantization with specific weight added to each data.
The above recording device generates and records quantization recording data by using the above various methods. The serial-system recording device causes a recording head to scan a recording medium in a main-scan direction to perform the recording of one line, moves the recording medium for a predetermined amount in a sub-scan direction after the one-line recording terminates, and again causes the recording head to scan the recording medium. Although the details will be explained later, the serial-system recording device repeats the above operation to perform the recording. On the other hand, the line-system recording device has recording means arranged entirely along a width direction of the recording medium. Thus, the line-system recording device performs the recording by sequentially scanning the recording medium.
However, in such a conventional quantization method and the recording device using the conventional quantization method, there are following inconveniences.
In the dither method, the number of gradations being representable is limited according to a size of a basic matrix. Further, since the gradation is represented by repeating the basic matrices, it happens according to a kind of an original image pattern that periodical unevenness called texture appears on the recording image at a period synchronous with a period of the basic matrix. Such the periodical unevenness appears particularly on a halftone image to be output, as the serious inconvenience. For these reasons, generally speaking, the dither method is the quantization method not suitable for a photographic natural image.
To cope with the above inconvenience, several proposals to attempt to improve quality of the halftone image in the dither method have been presented. For example, as disclosed in Japanese Patent Application Laid-Open No. 56-146361, it has been proposed that a dither mask pattern is made different for each color. However, such a technique disclosed in Japanese Patent Application Laid-Open No. 56-146361 is merely the method in which, e.g., a dither matrix is formed simply by using random numbers, and thus does not reach a level to reproduce the sufficient halftone image quality.
On the other hand, in the ED method, since the error data between the original image density and the output density is spatially calculated, the number of gradations which is limited due to the matrix size in the dither method is not limited. Further, an output image faithful to the original image density can be reproduced. Therefore, as compared with the dither method, faithful halftone recording can be realized. Furthermore, in the ED method, compatibility of the number of gradations with resolution which can not be realized in the dither method can be realized. For these reasons, in case of outputting a photographic image of which needs have been increased particularly in recent years, the ED method has been generally accepted as a means capable of realizing an increase in the image quality as compared with the dither method. However, since the number of steps necessary to process one pixel in the ED method is significantly larger than that in the dither method, a remarkable process time is necessary in the ED method. The needs for the high-quality image has caused a rapid increase in the recording resolution of the recording device. However, the number of pixels to be processed has further increased in proportion to a square of the increase in the recording resolution. Therefore, a situation that, in the ED method, a process speed dominates a recording speed of the recording device occurs.
In case of reproducing the halftone gradation by using the recording device, it is strongly expected to establish a quantization technique which has both high speed performance of the dither method and satisfactory halftone reproduction performance of the ED method. As one proposal to realize such the technique, e.g., U.S. Pat. No. 5,111,310 discloses a quantization technique in which, as the feature of the high-speed process in quantization control equivalent to the process speed in the conventional dither method is maintained by using a sufficiently large dither matrix, a pattern representing a spatial frequency characteristic called a blue noise is laid out to the matrix to realize such the halftone reproduction as in the ED method. It should be noted that the above method is called as a blue noise dither method hereinafter. The details of the blue noise dither method are disclosed in many publications, e.g., Robert Ulichneey, “Digital Halftoning,” The MIT Press Cambridge, Massachusetts London, England, and the like. The blue noise dither method is the method to perform the quantization by using the dither matrix which reduces a feeling of noise on the output image by suppressing a power spectrum of a low frequency component to which human eyes are most sensitive.
However, in such the blue noise dither method, the dither matrix is structured such that the blue noise pattern is laid out each color. Therefore, in a color image recording device, it happens that a mixed (or composite) color recorded on the basis of a result of a blue noise dither process performed for each color does not form or produce the blue noise pattern, whereby there is a drawback that the color image quality becomes degraded.
SUMMARY OF THE INVENTION
The present invention has been made in consideration of the above-described conventional problems, and an object of the present invention is to provide image process apparatus and method in which color image data of at least two colors are quantized by using different masks. In the apparatus and method, the quantization is performed by using
Canon Kabushiki Kaisha
Fitzpatrick ,Cella, Harper & Scinto
Grant II Jerome
Worku Negussie
LandOfFree
Image process apparatus and image process method 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 process apparatus and image process method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image process apparatus and image process method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2832410