CN102523364A - Document image strike-through eliminating method and system - Google Patents
Document image strike-through eliminating method and system Download PDFInfo
- Publication number
- CN102523364A CN102523364A CN2011103943422A CN201110394342A CN102523364A CN 102523364 A CN102523364 A CN 102523364A CN 2011103943422 A CN2011103943422 A CN 2011103943422A CN 201110394342 A CN201110394342 A CN 201110394342A CN 102523364 A CN102523364 A CN 102523364A
- Authority
- CN
- China
- Prior art keywords
- file
- picture
- pixel
- threshold value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a document image strike-through eliminating method and system and relates to the field of strike-through eliminating techniques in image processing. The method comprises the steps of: firstly finding out the character stroke area (foreground area) and non-character stroke area (background area) of a document image according to the global or local distribution characteristic of the document image, then conducting histogram statistic analysis to the background area to determine a strike-through eliminating threshold, and finally adjusting the document image according to the strike-through eliminating threshold to achieve the goal of eliminating strike-through. The method and the system disclosed by the invention have the advantages that the strike-through can be eliminated by self-adaptively determining the strike-through eliminating threshold, the strike-through at different levels can be eliminated, the eliminating effect to dark and light strike-through is good, the strike-through eliminating processing efficiency of the document image is greatly improved and the labor and the labor cost are saved.
Description
Technical field
Technical field is removed in the print through that the present invention relates in the image processing, is specifically related to a kind of file and picture print through sweep-out method and system.
Background technology
The digitized first step of paper document is scanning, and the black literal at the paper document back side can form a kind of blurred picture through scanner in the process of scanning, and this blurred picture just is called print through.The gray value of print through is generally between character image and background image.Traditional print through sweep-out method mainly relies on manually carries out, and not only time-consuming but also require great effort, efficient is very low.
Present print through sweep-out method mainly is to utilize the color range function of Photoshop to accomplish through operating personnel, and the basic procedure of its way is following: at first utilize Photoshop to open pending image, start the color range order; Regulate the function panel of color range then; It is the panchromatic passage of channel selecting; And the white slide block (from right to left in the moving process, the print through meeting in the image reduces gradually at slide block) that slidingly inputs the rightmost side in the color range from right to left till the print through complete obiteration, is clicked at last and is confirmed; Be applied to image, preserve image.Though the method can be removed print through preferably, waste time and energy, need the operating personnel that are familiar with Photoshop in a large number to accomplish.Therefore a kind of adaptive file and picture print through removal method of automation is very necessary.
The application for a patent for invention that application number is 200310124357.2, publication number is CN1512760A, name is called " removing the image processing apparatus and the method thereof that read the print through that two-sided or overlapping original copy produces " discloses a kind of through proofreading and correct to remove at the processing unit that reads the print through that two-sided or overlapping original copy produces.This patent is with any image inversion in direct picture or the reverse image; Detect the direct picture and the back side image of counter-rotating; Perhaps the back side image of counter-rotating and the position between the direct picture concern, and come correcting image so that remove the image print through according to detected position relation through computing.Though the method in this invention can be removed the most print throughs in the image preferably; But owing to just extracted the high brightness component of front, back side image when concern in the position of detected image; Do not pay attention to for being lower than the low-light level component of setting brightness value; So can not realize comparatively ideal removing print through through this method, especially handle all the more so with regard to the print through of lower image for brightness itself.
Summary of the invention
To the needs in defective that exists in the prior art and the practical application; The object of the present invention is to provide a kind of file and picture print through sweep-out method and system; The print through process automation is gone in realization, and the file and picture dissimilar, that the print through degree is different adaptive gone print through.
For realizing above-mentioned purpose, the technical scheme that the present invention adopts is: pending file and picture is selected in (1), finds out the strokes of characters zone and the non-legible stroke zone of file and picture;
(2) statistics with histogram is carried out in said non-legible stroke zone, and through histogram analysis being confirmed print through threshold value T;
(3) according to going print through threshold value T that said file and picture is adjusted to remove print through, the pixel value that is higher than the pixel of print through threshold value T is set to 255, does the functional relation conversion with what pixel value was lower than print through threshold value T.
Further; Aforesaid a kind of file and picture print through sweep-out method; In the step (1), according to the local distribution characteristic of said file and picture, perhaps overall distribution characteristics is found out the strokes of characters zone and the non-legible stroke zone of file and picture; Wherein, said local distribution characteristic comprises local mean value and variance.
Further, aforesaid a kind of file and picture print through sweep-out method, in the step (1), the concrete steps of finding out strokes of characters zone and non-legible stroke zone according to the local distribution characteristic of said file and picture are:
A) pixel value of each pixel in the said file and picture of calculating is the center with any pixel, and the pixel value of establishing this pixel is g (x; Y); Calculate the pixel value of all pixels in this center p * q scope average m (x, y) with variance s (x, y); 30≤p≤240,30≤q≤240 wherein;
B) according to average m (x, y) with variance s (x, y) calculate file and picture pixel value threshold value t (x, y), computing formula is:
Wherein, 0.2≤k
1≤0.5, R representes the maximum of standard deviation;
C) according to said threshold value t (x, y) each pixel to file and picture carries out threshold process, with pixel value little or equal threshold value t (x, pixel y) is set to the strokes of characters zone, (x, pixel y) are set to non-legible stroke zone greater than threshold value t.
Further, aforesaid a kind of file and picture print through sweep-out method, in the step a), the preferred value of p is 60, the preferred value of q is 60.
Further; Aforesaid a kind of file and picture print through sweep-out method, in the step (1), find out strokes of characters zone and non-legible stroke zone according to the local distribution characteristic of said file and picture after; Expansion process is carried out in said strokes of characters zone, and the number of times of expansion process is 1～3 time.
Further, aforesaid a kind of file and picture print through sweep-out method, in the step (2), statistics with histogram is carried out in non-legible stroke zone, and be through the concrete steps of histogram analysis being confirmed the print through threshold value:
I) calculate the number nTotal of non-legible stroke zone interior pixel;
Ii) set up the histogram in non-legible stroke zone;
Iii) travel through histogram from right to left, with in the histogram first less than k
2* the pairing pixel value of the post of nTotal is confirmed as print through threshold value T; Wherein, k
2Be empirical value, 0.0001≤k
2≤0.001.
Further again, aforesaid a kind of file and picture print through sweep-out method in the step (3), is selected linear function or nonlinear function when doing the functional relation conversion.
A kind of file and picture print through scavenge system based on above-mentioned file and picture print through sweep-out method comprises
Regional search device: be used for selected pending file and picture, find out the strokes of characters zone and the non-legible stroke zone of file and picture;
The threshold calculations device: be used to confirm the print through threshold value, this device comprises:
Number of pixels computing unit: the number nTotal that is used to calculate non-legible stroke image-region interior pixel;
Histogram is set up the unit: the histogram that is used to set up non-legible stroke zone;
Go the print through threshold value to confirm the unit: be used for traveling through from right to left histogram, and with in the histogram first less than k
2* the pairing pixel value of the post of nTotal is confirmed as print through threshold value T, k
2Be empirical value, 0.0001≤k
2≤0.001;
Image processing apparatus: be used for according to going print through threshold value T that said file and picture is adjusted the removing print through, the pixel value that is higher than print through threshold value T pixel is set to 255, does the functional relation conversion with what pixel value was lower than print through threshold value T.
Further, aforesaid a kind of file and picture print through scavenge system, said regional search device also comprises
Expansion process unit: be used for expansion process is carried out in the strokes of characters zone of confirming.
Effect of the present invention is: the present invention can the required threshold value of adaptive definite removal print through through the background area of file and picture being carried out statistical analysis, and according to this threshold value image is adjusted, and reaches the purpose of removing print through.Through these method and system, realized going the automation of print through process, and can improve treatment effeciency, labor savings and the labour cost of file and picture greatly the adaptive removal print through of file and pictures dissimilar, that the print through degree is different.
Description of drawings
Fig. 1 is the structured flowchart of a kind of file and picture print through of the present invention scavenge system;
Fig. 2 is the flow chart of a kind of file and picture print through of the present invention sweep-out method;
Fig. 3 is the neighborhood sketch map of pixel when seeking the strokes of characters zone in the embodiment;
Fig. 4 is the original document image before the print through that do not go among the embodiment;
Fig. 5 is the partial enlarged drawing of embodiment Fig. 4;
Fig. 6 is the strokes of characters area image among the embodiment;
Fig. 7 is the strokes of characters area image of process expansion process among the embodiment;
Fig. 8 is the histogram in non-legible stroke zone among the embodiment;
Fig. 9 is through the adjustment of linear function conversion figure as a result among the embodiment;
Figure 10 is the partial enlarged drawing among Fig. 9.
Embodiment
Below in conjunction with Figure of description and embodiment the present invention is done further detailed description.
Fig. 1 shows the structured flowchart of a kind of file and picture print through of the present invention scavenge system, by finding out that this system mainly comprises with lower device among the figure:
Regional search device 11: be used for selected pending file and picture, find out the strokes of characters zone and the non-legible stroke zone of file and picture;
Threshold calculations device 12: be used to confirm the print through threshold value
Image processing apparatus 13: be used for according to going print through threshold value T that said file and picture is adjusted the removing print through, the pixel value that is higher than print through threshold value T pixel is set to 255, does the functional relation conversion with what pixel value was lower than print through threshold value T.
Wherein, Regional search device 11 also comprises the expansion process unit that the strokes of characters zone is carried out expansion process; Because the strokes of characters zone of only finding out according to the local distribution characteristic maybe be not too accurate; It is carried out expansion process can guarantee covering strokes of characters area pixel as much as possible, make to the statistics in non-legible stroke zone the least possible as to receive the influence in strokes of characters zone.
Number of pixels computing unit: the number nTotal that is used to calculate non-legible stroke image-region interior pixel;
Histogram is set up the unit: the histogram that is used to set up non-legible stroke zone;
Go the print through threshold value to confirm the unit: be used for traveling through from right to left histogram, and with in the histogram first less than k
2* the pairing pixel value of the post of nTotal is confirmed as print through threshold value T, k
2Be empirical value, its span is [0.0001～0.001].
Passing threshold calculation element 12 is confirmed print through threshold value T; Image processing apparatus 12 is adjusted file and picture according to this print through threshold value T; The pixel that is higher than threshold value T for pixel value; The pixel value of this point is set to 255, carries out conversion process for what pixel value was lower than threshold value T according to certain functional relation (linear function or nonlinear function).
Fig. 2 shows the flow chart based on a kind of file and picture print through sweep-out method of file and picture print through scavenge system among Fig. 1, and this method comprises following step:
Step S21: confirm strokes of characters zone and non-legible stroke zone;
Selected pending file and picture confirms that the strokes of characters zone and the non-legible stroke of pending file and picture is regional.The strokes of characters zone is meant the pairing paper document of file and picture front strokes of characters institute region covered, and other are non-legible stroke zone (background area).Because print through is a kind of relatively fuzzyyer image; The strokes of characters zone is then comparatively sharp keen; Therefore can be regional with non-legible stroke according to the strokes of characters zone that the local distribution characteristic or the overall distribution characteristics of file and picture found out in the file and picture, the approximate region of strokes of characters is extracted.
For example can utilize the local mean value and the variance characteristic of file and picture to realize that concrete process is following:
A) pixel value of each pixel in the calculating file and picture is the center with any pixel, and the pixel value of establishing this pixel is g (x; Y), the calculation window size is that (x is y) with variance s (x for the average m of the pixel value of all pixels in p * q scope; Y); 30≤p≤240,30≤q≤240 wherein, p, the equal value of q are 60 in this embodiment;
B) according to average m (x, y) with variance s (x, y) calculate file and picture pixel value threshold value t (x, y), computing formula is:
k
1Be coefficient, its span is [0.2～0.5]; , R representes the maximum of standard deviation, for grey image R=128;
C) according to threshold value t (x, y) each pixel to file and picture carries out threshold process, processing formula is following:
O (x, the pixel value after y) expression is handled, this step be about to pixel value little or equal threshold value t (x, y) confirm as the strokes of characters zone, greater than threshold value t (x, y) confirm as non-legible stroke zone (background area).As shown in Figure 3, this figure is to be the center with center pixel (pixel value is 53), the picture of 60 * 60 window size pixel; The average of all pixel values in 60 * 60 scopes is 203.30, and variance is 59.26, is 170 through the pixel threshold of calculating this point; This pixel value 53＜170, therefore, this point is at o (x; Y) pairing pixel value is 0 in, i.e. this is the strokes of characters zone for representative.
Because the strokes of characters zone through above-mentioned steps is found out maybe be not too accurate; In actual process; Can also carry out expansion process (expansion is meant the morphology expansion process of bianry image, generally expands to get final product for 1～3 time) to the strokes of characters zone, the number of times of expansion process is an image and deciding; Carry out expansion process and can guarantee covering strokes of characters area pixel as much as possible, make to the statistics in non-legible stroke zone (background area) the least possible as to receive the influence in strokes of characters zone.
Step S22: non-legible stroke zone is carried out statistics with histogram and confirmed the print through threshold value;
Statistics with histogram is carried out in non-legible stroke zone, and through histogram analysis being confirmed print through threshold value T, concrete steps are following:
I) calculate the number nTotal of non-legible stroke zone interior pixel;
Ii) set up the histogram in non-legible stroke zone, and histogram is carried out smoothing processing;
Iii) travel through histogram from right to left, with in the histogram first less than k
2* nTotal (k
2Be empirical value) post, write down the pairing pixel value of this post, this pixel value is confirmed as print through threshold value T, k
2Span is [0.0001～0.001].
Step S23: according to going the print through threshold value that file and picture is handled.
According to going the print through threshold value said file and picture is adjusted to remove print through; Concrete adjustment is that the pixel value that pixel value is higher than print through threshold value T is set to 255; What will be lower than T carries out conversion according to certain functional relation, can select different functions that image is handled, for example linear function, nonlinear function etc.; Adopt different functions that image is handled, the image of final adjustment can be different.
Below in conjunction with concrete embodiment the present invention is carried out further detailed explanation.
Embodiment
Fig. 4 is the file and picture that has print through for handling in the present embodiment; Fig. 5 is the partial enlarged drawing of Fig. 4; By figure in can find out that the print through of this figure is comparatively serious, adopt method and system of the present invention to remove print through after, its design sketch such as Fig. 9 and shown in Figure 10; Can find out, remove the print through in the file and picture well through method of the present invention.Carry out detailed explanation in the face of the process of removing print through down:
At first,, adopt average and variance characteristic in the present embodiment according to the local distribution characteristic of Fig. 4, the calculating pixel threshold value, and according to this threshold value each pixel is handled, the strokes of characters of finding out file and picture is regional.As shown in Figure 6, black region is the strokes of characters zone, and white portion is non-legible stroke zone.Because the strokes of characters zone of only finding out according to the local distribution characteristic is not too accurate; In order to guarantee covering strokes of characters area pixel as much as possible; Make to the least possible influence that receives the strokes of characters zone of the statistics of background area; Expansion process is carried out in the strokes of characters zone, and the number of times of expansion process is an image and deciding, and Fig. 7 is for to carry out expansion process result once to Fig. 6 Chinese words stroke zone.After having confirmed strokes of characters zone and non-legible stroke zone, statistics with histogram (Fig. 8) is carried out in non-legible stroke zone, and confirm the print through threshold value through histogram is analyzed according to step S22, the qualified print through threshold value of going is 170 in the present embodiment.The print through threshold value of going that last basis obtains is adjusted image; Be changed to 255 with what pixel value was higher than the print through threshold value; What be lower than the print through threshold value carries out conversion process according to linearity or nonlinear function; Adopt different functions that image is handled, the result of final adjustment understands different, and Fig. 9 is the adjustment result who carries out linear transformation.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, belong within the scope of claim of the present invention and equivalent technology thereof if of the present invention these are revised with modification, then the present invention also is intended to comprise these changes and modification interior.
Claims (9)
1. file and picture print through sweep-out method may further comprise the steps:
(1) selected pending file and picture, strokes of characters zone and the non-legible stroke of finding out file and picture are regional;
(2) statistics with histogram is carried out in said non-legible stroke zone, and through histogram analysis being confirmed print through threshold value T;
(3) according to going print through threshold value T that said file and picture is adjusted to remove print through, the pixel value that is higher than the pixel of print through threshold value T is set to 255, does the functional relation conversion with what pixel value was lower than print through threshold value T.
2. a kind of file and picture print through sweep-out method as claimed in claim 1; It is characterized in that: in the step (1); Local distribution characteristic according to said file and picture; Perhaps overall distribution characteristics is found out the strokes of characters zone and the non-legible stroke zone of file and picture, and wherein, said local distribution characteristic comprises local mean value and variance.
3. a kind of file and picture print through sweep-out method as claimed in claim 2 is characterized in that: in the step (1), the concrete steps of finding out strokes of characters zone and non-legible stroke zone according to the local distribution characteristic of said file and picture are:
A) pixel value of each pixel in the said file and picture of calculating is the center with any pixel, and the pixel value of establishing this pixel is g (x; Y); Calculate the pixel value of all pixels in this center p * q scope average m (x, y) with variance s (x, y); 30≤p≤240,30≤q≤240 wherein;
B) according to average m (x, y) with variance s (x, y) calculate file and picture pixel value threshold value t (x, y), computing formula is:
Wherein, 0.2≤k1≤0.5, R representes the maximum of standard deviation;
C) according to said threshold value t (x, y) each pixel to file and picture carries out threshold process, with pixel value little or equal threshold value t (x, pixel y) is set to the strokes of characters zone, (x, pixel y) are set to non-legible stroke zone greater than threshold value t.
4. like the said a kind of file and picture print through sweep-out method of claim 3, it is characterized in that: in the step a), the preferred value of p is 60, and the preferred value of q is 60.
5. like claim 2 or 3 said a kind of file and picture print through sweep-out methods; It is characterized in that: in the step (1); After finding out strokes of characters zone and non-legible stroke zone according to the local distribution characteristic of said file and picture; Expansion process is carried out in said strokes of characters zone, and the number of times of expansion process is 1～3 time.
6. a kind of according to claim 1 file and picture print through sweep-out method is characterized in that: in the step (2), statistics with histogram is carried out in non-legible stroke zone, and through the concrete steps of histogram analysis being confirmed the print through threshold value be:
I) calculate the number nTotal of non-legible stroke zone interior pixel;
Ii) set up the histogram in non-legible stroke zone;
Iii) travel through histogram from right to left, with in the histogram first less than k
2* the pairing pixel value of the post of nTotal is confirmed as print through threshold value T; Wherein, k
2Be empirical value, 0.0001≤k
2≤0.001.
7. the sweep-out method of a kind of file and picture print through as claimed in claim 1 is characterized in that: in the step (3), select linear function or nonlinear function when doing the functional relation conversion.
8. a file and picture print through scavenge system comprises
Regional search device: be used for selected pending file and picture, find out the strokes of characters zone and the non-legible stroke zone of file and picture;
The threshold calculations device: be used to confirm the print through threshold value, this device comprises:
Number of pixels computing unit: the number nTotal that is used to calculate non-legible stroke image-region interior pixel;
Histogram is set up the unit: the histogram that is used to set up non-legible stroke zone;
Go the print through threshold value to confirm the unit: be used for traveling through from right to left histogram, and with in the histogram first less than k
2* the pairing pixel value of the post of nTotal is confirmed as print through threshold value T, k
2Be empirical value, 0.0001≤k
2≤0.001;
Image processing apparatus: be used for according to going print through threshold value T that said file and picture is adjusted the removing print through, the pixel value that is higher than print through threshold value T pixel is set to 255, does the functional relation conversion with what pixel value was lower than print through threshold value T.
9. a kind of file and picture print through scavenge system as claimed in claim 8, it is characterized in that: said regional search device also comprises
Expansion process unit: be used for expansion process is carried out in the strokes of characters zone of confirming.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103943422A CN102523364A (en) | 2011-12-02 | 2011-12-02 | Document image strike-through eliminating method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103943422A CN102523364A (en) | 2011-12-02 | 2011-12-02 | Document image strike-through eliminating method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102523364A true CN102523364A (en) | 2012-06-27 |
Family
ID=46294149
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011103943422A Pending CN102523364A (en) | 2011-12-02 | 2011-12-02 | Document image strike-through eliminating method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102523364A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105245756A (en) * | 2015-09-28 | 2016-01-13 | 珠海奔图电子有限公司 | Image processing method and system |
| CN105335935A (en) * | 2014-06-11 | 2016-02-17 | 富士通株式会社 | Image processing apparatus and method |
| CN105741272A (en) * | 2016-01-25 | 2016-07-06 | 上海中信信息发展股份有限公司 | Method for removing osmotic writing on back surface of document image |
| CN105898104A (en) * | 2015-01-04 | 2016-08-24 | 陈于中 | Method for substracting backside content revealed from scanned image |
| CN106023090A (en) * | 2016-03-17 | 2016-10-12 | 陈于中 | Method of eliminating displayed back side contents in shot page image |
| CN108513039A (en) * | 2017-02-28 | 2018-09-07 | 佳能株式会社 | Image processing apparatus, image processing method and recording medium |
| CN110619608A (en) * | 2018-06-20 | 2019-12-27 | 佳能株式会社 | Image processing apparatus, control method thereof, and storage medium |
| CN112784850A (en) * | 2019-11-04 | 2021-05-11 | 珠海金山办公软件有限公司 | Method and device for removing penetrating print of notes |
| CN113784009A (en) * | 2021-09-10 | 2021-12-10 | 北京航星永志科技有限公司 | Paper text image processing method and device and electronic equipment |
| WO2023046112A1 (en) * | 2021-09-26 | 2023-03-30 | 维沃移动通信有限公司 | Document image enhancement method and apparatus, and electronic device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1419679A (en) * | 2000-03-14 | 2003-05-21 | 英特尔公司 | Estimating text color and segmentation of images |
| CN1936959A (en) * | 2005-09-23 | 2007-03-28 | 西门子公司 | Method for combining two images based on eliminating background pixels from one of the images |
| CN101231756A (en) * | 2008-01-30 | 2008-07-30 | 安防科技（中国）有限公司 | Method and apparatus for detecting moving goal shade |
-
2011
- 2011-12-02 CN CN2011103943422A patent/CN102523364A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1419679A (en) * | 2000-03-14 | 2003-05-21 | 英特尔公司 | Estimating text color and segmentation of images |
| CN1936959A (en) * | 2005-09-23 | 2007-03-28 | 西门子公司 | Method for combining two images based on eliminating background pixels from one of the images |
| CN101231756A (en) * | 2008-01-30 | 2008-07-30 | 安防科技（中国）有限公司 | Method and apparatus for detecting moving goal shade |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335935A (en) * | 2014-06-11 | 2016-02-17 | 富士通株式会社 | Image processing apparatus and method |
| CN105335935B (en) * | 2014-06-11 | 2018-03-16 | 富士通株式会社 | Image processing apparatus and method |
| CN105898104A (en) * | 2015-01-04 | 2016-08-24 | 陈于中 | Method for substracting backside content revealed from scanned image |
| US10497080B2 (en) | 2015-09-28 | 2019-12-03 | Zhuhai Seine Technology Co., Ltd. | Method and apparatus for image processing |
| WO2017054637A1 (en) * | 2015-09-28 | 2017-04-06 | 珠海赛纳打印科技股份有限公司 | Image processing method and system |
| CN105245756B (en) * | 2015-09-28 | 2018-05-29 | 珠海奔图电子有限公司 | Image processing method and system |
| CN105245756A (en) * | 2015-09-28 | 2016-01-13 | 珠海奔图电子有限公司 | Image processing method and system |
| CN105741272A (en) * | 2016-01-25 | 2016-07-06 | 上海中信信息发展股份有限公司 | Method for removing osmotic writing on back surface of document image |
| CN106023090A (en) * | 2016-03-17 | 2016-10-12 | 陈于中 | Method of eliminating displayed back side contents in shot page image |
| CN108513039A (en) * | 2017-02-28 | 2018-09-07 | 佳能株式会社 | Image processing apparatus, image processing method and recording medium |
| CN108513039B (en) * | 2017-02-28 | 2020-08-21 | 佳能株式会社 | Image processing apparatus, image processing method, and recording medium |
| US10999473B2 (en) | 2017-02-28 | 2021-05-04 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and recording medium to correct characteristic amount of target pixel using brightness at two pixels in set target area |
| CN110619608A (en) * | 2018-06-20 | 2019-12-27 | 佳能株式会社 | Image processing apparatus, control method thereof, and storage medium |
| CN110619608B (en) * | 2018-06-20 | 2023-11-14 | 佳能株式会社 | Image processing apparatus, control method thereof, and storage medium |
| CN112784850A (en) * | 2019-11-04 | 2021-05-11 | 珠海金山办公软件有限公司 | Method and device for removing penetrating print of notes |
| CN113784009A (en) * | 2021-09-10 | 2021-12-10 | 北京航星永志科技有限公司 | Paper text image processing method and device and electronic equipment |
| CN113784009B (en) * | 2021-09-10 | 2022-10-18 | 北京航星永志科技有限公司 | Paper text image processing method and device and electronic equipment |
| WO2023046112A1 (en) * | 2021-09-26 | 2023-03-30 | 维沃移动通信有限公司 | Document image enhancement method and apparatus, and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102523364A (en) | Document image strike-through eliminating method and system | |
| CN102831591B (en) | Gaussian filter-based real-time defogging method for single image | |
| CN103747213B (en) | A kind of real-time defogging method of the Traffic Surveillance Video based on moving target | |
| CN103761720B (en) | Image defogging method and image demister | |
| TWI431559B (en) | A vehicle registration-plate detecting method and system thereof | |
| EP2568438A2 (en) | Image defogging method and system | |
| CN102202163B (en) | Adaptive enhancement method and device for monitored video | |
| CN103065334B (en) | A kind of color cast detection based on hsv color space, bearing calibration and device | |
| CN106251301A (en) | A kind of single image defogging method based on dark primary priori | |
| CN105225210A (en) | A kind of self-adapting histogram based on dark strengthens defogging method capable | |
| US8213735B2 (en) | Methods and apparatus for performing image binarization | |
| CN104794688A (en) | Single image defogging method and device based on depth information separation sky region | |
| CN105374015A (en) | Binary method for low-quality document image based on local contract and estimation of stroke width | |
| CN102411775B (en) | Method and system for correcting brightness of text image | |
| CN103996178A (en) | Sand and dust weather color image enhancing method | |
| CN102611828A (en) | Real-time enhanced processing system for foggy continuous video image | |
| CN107705254B (en) | City environment assessment method based on street view | |
| CN105046677A (en) | Enhancement processing method and apparatus for traffic video image | |
| CN101710425A (en) | Self-adaptive pre-segmentation method based on gray scale and gradient of image and gray scale statistic histogram | |
| CN105023256A (en) | Image defogging method and system | |
| CN108564538A (en) | A kind of image based on ambient light difference goes haze method and system | |
| CN109741273A (en) | A kind of mobile phone photograph low-quality images automatically process and methods of marking | |
| US7668394B2 (en) | Background intensity correction of a scan of a document | |
| CN101315664A (en) | Text image preprocessing method for character recognition | |
| CN104657969B (en) | A kind of image shading minimizing technology and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20120627 |
|
| C20 | Patent right or utility model deemed to be abandoned or is abandoned |