Image analysis – Image enhancement or restoration
Reexamination Certificate
1999-01-14
2002-06-25
Lee, Thomas D. (Department: 2624)
Image analysis
Image enhancement or restoration
C382S261000
Reexamination Certificate
active
06411741
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus for enhancing the quality of characters (line image components) included in an image formed, for example, by a digital color electronic copying machine for forming a color image.
In general, in the field of image forming apparatus, such as a digital PPC (Plain Paper Copier), for enhancing the quality of characters (line image components) included in an image, there is known a method of emphasizing characters (line image components) by detecting a density gradient (BAT method) of a predetermined area on the image, discriminating a portion with a high gradient to be a character (line image component), and increasing the density of the discriminated portion, thereby emphasizing a contrast between the character (line image component) and the background portion and thus enhancing the image quality. There is also known a method wherein the above-mentioned discriminated portion is expanded in a main scan direction and a sub-scan direction and thus the contrast is emphasized.
However, with the above-mentioned discrimination technique of BAT method, the density of only a contour portion of the character (line image component) is increased and an inside portion of the character (line image component) is viewed to have a low density.
Recently, in most copying machines, a discrimination process is performed in an initial setting mode on an original on which a character image and a gradient image are mixed, and predetermined image processing is performed on the character image and gradient image. It is desired that of those images, the character image be subjected to a process for obtaining a clearer image even if the number of gradients decreases, and the gradient image be subjected to a process for obtaining a smoother image with a large number of gradients.
An area discrimination technique is thus indispensable for determining which individual pixels of an image input to the copying machine belong to which image portions.
Jpn. Pat. Appln. KOKOKU Publication No. 4-5305, etc. describes a prior-art area discrimination method in which a two-dimensional block of a predetermined size is prepared, a pixel value of a pixel of interest is compared with a pixel value of a peripheral pixel, and if a difference is greater than a predetermined value, the presence of an edge is determined and detected as a component of a character or a line.
Jpn. Pat. Appln. KOKOKU Publication No. 5-56067, Jpn. Pat. Appln. KOKOKU Publication No. 5-50187, Jpn. Pat. Appln. KOKOKU Publication No. 5-46789, etc. describe that absolute values of a difference between adjacent image components in a two-dimensional block of a predetermined size are found in a main scan direction and a sub-scan direction, and a sum of integrated values is compared with a predetermined threshold value, thereby discriminating attributes of the image components.
In such methods, in general, detection is made on the basis of the found characteristics amount under predetermined conditions. In most of such methods, an edge portion of a character alone is discriminated. However, in a case where a character with thick lines, such as a logo, is discriminated by a method of using an edge portion, if an emphasis process is performed, only an edge portion discriminated as a character is emphasized and a so-called “hollow state” occurs within the character.
Besides, there is an original on which an edge itself cannot easily be detected, like a pencil-written manuscript on which a character itself is thin. In such a case, most of characters are not discriminated and consequently characters obtained as outputs become difficult to read.
As has been described above, if the discrimination process using the character discrimination method based merely on the characteristic amount for detecting the so-called edge portion of the image is performed, as in the prior art, only the edge portion of the character as in the case of a logo including a thick line and the inner area of the character is not detected. As a result, the so-called “hollow state” occurs.
Moreover, there is an original, such as a pencil-written manuscript, on which a character itself is thin and the edge portion thereof is difficult to detect. Since most of characters are not discriminated in this case, most characters are output in a blurring state. As a result, output characters become difficult to read.
If the above problems are to be solved by the process of expanding discrimination signals, the scale of hardware increases undesirably due to an increase in the number of line memories.
BRIEF SUMMARY OF THE INVENTION
The object of the present invention is to provide an image processing apparatus wherein not only an edge portion but also an inside portion of a character (line image component) is discriminated and the quality of the character (line image component) is enhanced.
According to the present invention, in order to achieve the object, there is provided an image processing apparatus comprising: characteristic amount calculation means for calculating a characteristic amount on the basis of an input image signal; and discrimination generation means for generating a discrimination signal for discriminating attributes of the input image signal, on the basis of the characteristic amount calculated by the characteristic amount calculation means and the input image signal, wherein an image process is switched in accordance with the discrimination signal generated by the discrimination generation means.
According to the present invention, in order to achieve the object, there is provided an image processing apparatus comprising: control means for outputting a control signal for switching calculation conditions for characteristic amounts and discrimination conditions for attributes in accordance with a set document mode; characteristic amount calculation means for calculating a characteristic amount of an input image signal on the basis of the calculation condition switched by the control signal output from the control means; first multiplication means for multiplying the characteristic amount calculated by the characteristic amount calculation means with a predetermined coefficient; second multiplication means for multiplying the input image signal with a predetermined coefficient; attribute discrimination means for generating a discrimination signal for discriminating attributes of the input image signal on the basis of an output from the first multiplication means and an output from the second multiplication means, under the attribute discrimination condition switched by the control signal from the control means; and output means for outputting an image signal which has been subjected to an image process according to the discrimination signal from the attribute discrimination means.
According to the present invention, in order to achieve the object, there is provided an image processing apparatus comprising: control means for effecting switching between an image signal for image processing and an image signal for pre-processing; processing means for finding a cumulative frequency of a signal value of the image signal for pre-processing, which is switched and input by the control means; characteristic amount calculation means for calculating a characteristic amount of the image signal for image processing, which is switched and input by the control means, on the basis of the cumulative frequency found by the processing means; first multiplication means for multiplying the characteristic amount calculated by the characteristic amount calculation means with a predetermined coefficient; second multiplication means for multiplying the image signal for image processing, which is switched and input by the control means, with a predetermined coefficient; attribute discrimination means for discriminating attributes of the input image signal for image processing, on the basis of an output from the first multiplication means and an output from the second multiplication means, with reference to the cumulative freque
Hamamura Naoko
Kawakami Haruko
Brinich Stephen
Foley & Lardner
Kabushiki Kaisha Toshiba
Lee Thomas D.
LandOfFree
Image processing apparatus 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 processing apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2944819