Image processing apparatus

Image analysis – Image segmentation – Distinguishing text from other regions

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C382S156000, C358S452000, C358S462000

Reexamination Certificate

active

06240206

ABSTRACT:

FIELD OF THE INVENTION
The present invention relates to an image processing apparatus, for use in a scanner, a digital copying machine, a facsimile, or other devices, which recognizes a character region, a photographic region, and a spot region of an image obtained by scanning a document so as to carry out an optimum process for each region according to its feature.
BACKGROUND OF THE INVENTION
It has been a problem of a conventional image processing apparatus that the sharpness or an input level of an image input section composed of an input device such as CCD (Charge Coupled Device) varies depending on individual apparatuses. For this reason, when shipping image processing apparatuses, if a region judgement (judging whether a region is a character region, a spot region, and a photographic region) using a reference document is carried out with respect to individual apparatuses, the result of the judgement is not coincident with respective ratio specified in the reference document. Here, the reference document includes a character region, a spot region, and a photographic region at such a specified ratio.
As a countermeasure, in order to compensate the variations in the input devices of the image processing apparatuses, an improvement in the mechanical and dimensional accuracy of the input device and correction such as an adjustment of an input level (shading correction) are made. However, the improvement in the mechanical and dimensional accuracy involves a substantial increase in cost. Also, the shading correction has not reached a level where the variations in the input devices are completely corrected. This results from the fact that the accuracy of a reference white plate density and the accuracy of mounting the white plate are not high enough.
In order to improve an image quality of an image processed by the image processing apparatus, while still having the problem of variations in the input devices, it has been a focus of ongoing research to develop technology which allows (1) separation of the character, photographic, and spot regions of a document which is subjected to image processing and (2) an optimum process to be carried out according to the feature of each region.
For example, Japanese Unexamined Patent Publication No. 133159/1994 (Tokukaihei 6-133159) discloses an arrangement of an image processing apparatus for accurate detection of a spot region with a simple arrangement. According to this arrangement, first, a local change in image data of a region is detected with respect to a main-scanning direction and a sub-scanning direction, respectively. Secondly, it is determined whether the region is a spot region based on pixels, in the region, having detected changes above a predetermined value. Finally, a smoothing filter is applied to the spot region so as to prevent moire, and an MTF correction filter (edge enhancement filter) is adopted to a non-spot region so as to sharpen the image.
Despite the fact that the quality of region separation has been improving in the past years, there has been no improvement as to the correction of variations in input devices. This presents a problem that the image quality of an image that has been image-processed deteriorates since the accuracy of region separation greatly changes depending on the degree of variations in the input devices. That is to say, despite the fact that an optimum process is carried out for each of the separated regions, the image quality deteriorates due to low accuracy of the region separation.
Even in the arrangement disclosed in the afore-mentioned publication, the accuracy of region separation greatly changes depending on the degree of the variations in the input devices. Because this arrangement does not have means for adjusting the result of the region separation so as to compensate for change in an input device and the variation in the input devices, it presents a problem that a high image quality is achieved in one apparatus, while posing a problem of a coexisting region in another apparatus in which a spot region and a non-spot region coexist. Such a coexisting region causes that the resulting image is likely to have a poor image quality.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide an image processing apparatus wherein the accuracy of region separation can be improved so as to improve the image quality of a final image that has been image-processed even if sharpness or input levels of input devices vary depending on individual image processing apparatuses.
In order to achieve the above-mentioned object, an image processing apparatus of the present invention includes a region separating section, upon receiving a reference document, for recognizing whether each part of the reference document belong to a character region, a photographic region, or a spot region of the reference document so as to carry out region separation for the reference document, the character region, the photographic region, and the spot region being provided at a specified ratio,
an automatic adjusting section for adjusting a region separation state of the region separating section based on (1) a result of region separation for the reference document by the region separating section and (2) the specified ratio of the character region, the photographic region, and the spot region in the reference document.
According to the image processing apparatus having the described arrangement, when the region separating section carries out region separation for the reference document, the automatic adjusting section adjusts the state of region separation for the region separating section based on (1) the result of the region separation by the region separating section and (2) the ratio specified in the reference document. Here, if the automatic adjusting section adjusts the state of region separation in such a manner that the ratio of each region which is determined from the result of the region separation for the reference document by the region separating section is substantially equal to the preset ratio of the region of the reference document, it is possible to improve the accuracy of the region separation even if variations exist in input devices through which image data are inputted to the image processing apparatus, thereby permitting optimum image processing to be carried out for each region so as to improve the image quality of the final image. Further, it is also possible to make the final image to have a desired image quality if the preset region separation state is set as the user prefers.
For a fuller understanding of the nature and advantages of the invention, reference should be made to the ensuing detailed description taken in conjunction with the accompanying drawings.


REFERENCES:
patent: 4797738 (1989-01-01), Kashi et al.
patent: 5001576 (1991-03-01), Tanaka et al.
patent: 5086346 (1992-02-01), Fujisawa
patent: 5194966 (1993-03-01), Quardt et al.
patent: 5481380 (1996-01-01), Bestmann
patent: 5608819 (1997-03-01), Ikeuchi
patent: 5787195 (1998-07-01), Tsujimoto et al.
patent: 5867593 (1999-02-01), Fukuda et al.
patent: 4305693 (1993-01-01), None
patent: 710004A2 (1996-05-01), None
patent: 61-247164 (1986-11-01), None
patent: 6-133159 (1994-05-01), None

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

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.

Rate now

     

Profile ID: LFUS-PAI-O-2546918

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.