CN109584163B - Method for restoring original file of paper scrap - Google Patents
Method for restoring original file of paper scrap Download PDFInfo
- Publication number
- CN109584163B CN109584163B CN201811544621.0A CN201811544621A CN109584163B CN 109584163 B CN109584163 B CN 109584163B CN 201811544621 A CN201811544621 A CN 201811544621A CN 109584163 B CN109584163 B CN 109584163B
- Authority
- CN
- China
- Prior art keywords
- paper
- scrap
- pixel
- value
- scraps
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 239000011159 matrix material Substances 0.000 claims abstract description 110
- 239000013598 vector Substances 0.000 claims description 87
- 239000012634 fragment Substances 0.000 claims description 22
- 238000011084 recovery Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 3
- 239000010893 paper waste Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006260 foam Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Crushing And Pulverization Processes (AREA)
Abstract
The invention provides a method for restoring original documents of paper scraps. The original file restoration method of the paper scraps is characterized in that the paper scraps located on the leftmost side of the original file are determined firstly, then other paper scraps are sequentially arranged on the right side of the paper scraps located on the leftmost side according to the absolute distance matrix relation between the other paper scraps and the paper scraps located on the leftmost side of the original file from small to large of the absolute distance matrix, so that the original file is restored, or the left boundary and the right boundary of a restoration matrix located on the original file are determined firstly, the upper boundary and the lower boundary of the restoration matrix are determined according to the left boundary and the right boundary of the restoration matrix, finally, human vision is combined with a computer, and the paper scraps in an idle state are spliced by using a human-computer interaction interface, so that the whole original file is restored, a large amount of human resources are saved, the working efficiency is improved, and the restoration accuracy is high.
Description
Technical Field
The invention relates to the technical field of image restoration, in particular to a method for restoring original files of paper scraps.
Background
Paper shredder has become an indispensable equipment in daily office, the paper shredder is composed of a group of rotating blades, a paper comb and a driving motor, paper is fed from the middle of the blades which are engaged with each other, and is divided into a plurality of fine paper sheets so as to achieve the purpose of confidentiality, the paper shredding mode is the shape of the paper which is shredded by a paper shredding knife after being processed by the paper shredder, and the existing paper shredding mode comprises the following steps according to the composition mode of the paper shredding knife: crushed, granular, segmented, foamed, striped, filamentous, and the like. The shredding effect is the size of waste paper formed after paper is processed by a shredder, and is generally in millimeter (mm) units, the particle and foam effects are the best, the shredding effect is the second, and the strip and segment relative effects are worse. For example, 2 x 2mm security effect can cut a4 paper into 1500 small pieces, different paper shredders with different shredding effects can be selected according to actual needs, for example, the specifications of 4mm x 50mm, 4mm x 30mm and the like can be selected in the occasions where families and small offices do not involve security, and the computer printed documents need to be shredded into paper strips with the thickness of less than 3.8mm according to the minimum standard of destroyed data in the occasions where security is required, for highly confidential documents, a shredder capable of being cut vertically and horizontally is adopted, and the shredder with the shredding effect of the specification of 3mm x 3mm and less is preferably selected.
Many enterprises, scientific research institutions and military forces can destroy important documents and data by using paper shredders according to the confidential requirements, so that the splicing of the broken documents has important application in the fields of judicial evidence recovery, historical literature repair, military information acquisition and the like. Traditional concatenation is recovered work and is relied on the manual work to accomplish, and the rate of accuracy is high, but efficiency is very low, and the secondary damage that causes among the manual concatenation process can cause more serious consequence, and along with the development of computer technology, people try to develop the automatic splicing technique of shredded paper piece to improve the concatenation and recover efficiency.
In the existing automatic and semi-automatic restoration technology by using a computer, the restoration of the broken file is mostly carried out according to the characteristics of the geometrical shape of the fragments. However, for regular pictures broken by shredders from the same printed text document, the way of restoring them by extracting the geometric features of the fragments is not applicable. Of course, the method of performing natural language processing or semantic analysis on the characters in the fragments through machine learning or pattern recognition is too complicated. Therefore, the technology for splicing and restoring the shredded paper sheets needs to be designed by paying attention to the distribution situation of characters in the shredded paper sheets, the margin of the document, particularly the positions of the characters in the shredded paper sheets, and the characteristics of texture distribution, fit degree and the like of the edges of the shredded paper sheets.
Disclosure of Invention
The invention aims to provide a method for restoring original documents of paper scraps, which can save a large amount of human resources, improve the working efficiency and have high restoration accuracy.
In order to achieve the above object, the present invention provides a method for restoring an original document of a paper scrap, comprising the steps of:
step S1, obtaining a plurality of paper fragments corresponding to the original document and a plurality of images corresponding to the paper fragments, reading a gray value of each pixel in the plurality of images, and obtaining a pixel matrix corresponding to each paper fragment: a. thei＝(a1，a2，Λ，an) Wherein A isiPixel matrix corresponding to the ith scrap piece, a1For the 1 st column pixel gray value vector, a, of the ith scrapnSetting i as a positive integer and n as a positive integer larger than 2 for the pixel gray value vector of the nth column of the ith paper scrap;
step S2, processing each row of pixel gray value vectors in the pixel matrix corresponding to each paper scrap according to a binarization formula to obtain a binarization value of each component value in each row of pixel gray value vectors of each paper scrap;
the binarization formula is as follows:
wherein R isi n(k) Is the binary value of the kth component value in the n-th column pixel gray value vector of the ith wastepaper sheet, an(k) Setting k as a natural number more than 2 for the kth component value in the n-th column of pixel gray value vector of the ith scrap paper sheet;
step S3, determining that the ith paper scrap is located at the position of the original file according to the binary value of each component value in the first column to the Z column pixel gray value vector of the ith paper scrap, when the binary value of each component value in the first column to the Z column pixel gray value vector of the ith paper scrap is 1, determining that the ith paper scrap is located at the leftmost side of the original file, and setting Z as a natural number greater than 1, or,
judging the positions of the ith paper scrap and the jth paper scrap on the original file according to an absolute distance matrix between the binary value of each component value in the 1 st column of pixel gray value vectors of the ith paper scrap and the binary value of each component value in the nth column of pixel gray value vectors of the jth paper scrap, wherein when the absolute distance matrix is 0, the ith paper scrap is positioned at the leftmost side of the original file, the jth paper scrap is positioned at the rightmost side of the original file, and j is set as a positive integer;
and step S4, according to an absolute distance matrix between the binary value of each component value in the nth column of pixel gray value vectors of the ith paper scrap and the binary value of each component value in the 1 st column of pixel gray value vectors of other paper scrap, arranging other paper scrap on the right side of the ith paper scrap in sequence according to the relationship from small to large of the absolute distance matrix, and thus restoring the original file.
The original document is a printed text document on the same page, and the paper scraps are longitudinally cut paper scraps of the original document.
In the step S3, the binarized value of each component value in the 1 st column pixel gradation value vector of the ith paper-scrap and each of the n th column pixel gradation value vectors of the jth paper-scrapThe absolute distance matrix between the binarized values of the component values is: dij＝∑|Ri 1(k)-Rj n(k) L, wherein DijIs an absolute distance matrix, R, between the 1 st column pixel gray value vector of the ith paper scrap and the n th column pixel gray value vector of the jth paper scrapi 1(k) Is the binary value, R, of the kth component value in the 1 st column pixel gray value vector of the ith scrap sheetj n(k) A binarized value of a k-th component value in the n-th column pixel gradation value vector for the jth chip sheet.
In step S4, the absolute distance matrix between the binarized value of each component value in the n-th column pixel grayscale value vector of the i-th scrap piece and the binarized value of each component value in the 1-th column pixel grayscale value vector of the other scrap piece is: dio＝∑|Ri n(k)-Ro 1(k) L, wherein DioIs an absolute distance matrix, R, between the gray value vector of the nth column pixel of the ith paper scrap and the gray value vector of the 1 st column pixel of other paper scrapi n(k) Is the binary value, R, of the kth component value in the n-th column pixel gray-scale value vector of the ith scrap sheeto 1(k) The binary value of the kth component value in the 1 st column pixel gray value vector of the other paper scrap is obtained.
The invention also provides a method for restoring the original file of the paper scrap, which comprises the following steps:
step S1', obtaining a plurality of paper fragments corresponding to the original document and a plurality of images corresponding to the plurality of paper fragments, reading a gray value of each pixel in the plurality of images, and obtaining a pixel matrix corresponding to each paper fragment:
wherein, BkPixel matrix for the kth scrap piece, bm，n kIs the gradation value of the pixel of the mth row and nth column of the kth chip sheet, ln kFor the pixel gray value vector of the nth column of the kth chip sheet, hm kLet k be positive for the pixel gray value vector of the mth row of the kth fragmentThe integer, m and n are all positive integers greater than 2;
step S2', the gray value of each pixel in the pixel matrix corresponding to each shredded paper is processed according to the binarization formula, and the gray binarization value of each pixel of each shredded paper is obtained;
the binarization formula is as follows:
wherein x is 1, 2, ·, m; y is 1, 2,. cndot.n; b'x，y kA gray level binary value of the pixel of the x row and the y column of the k scrap paper sheet;
step S3', set the restoration matrix of the original file as
Wherein, Cf×gTo restore the matrix, Cf，gFor scraps of paper located in the f-th row and g-th column,/gFor scraps of paper in the g-th column, hfIs a scrap piece located in row f; record l1 kThe number of the pieces of paper which are all 1 is E1When E is1>f is then at E1Record l in one scrap piece2 kThe number of the pieces of paper which are all 1 is E2When E is2>f is then at E2Record l in one scrap piece3 kThe number of all 1 scraps was E3, up to EnIf f, then determine ln kAll 1EnThe individual scraps of paper are positioned in the first column of the restoration matrix, thus being positioned in l1The shredded paper of (1);
step S4', record ln kThe number of the pieces of paper which are all 1 is En', when En’>f is then at En' record in one scrap piecen-1 kThe number of the pieces of paper which are all 1 is En-1', when En-1’>f is then at En-1' record in one scrap piecen-2 kThe number of the pieces of paper which are all 1 is En-2', up to En-pIf f, then determine ln-p kAll 1En-p' one scrap piece is located in the last column of the recovery matrix, thus being located exactly ingSetting p as a positive integer smaller than n;
step S5' obtaining signals at l1C in (1)1，1Shredded paper and a sheet located ingC in (1)1，gPixel matrix of a chip sheet, record C1，1Paper scraps and C1，gThe method comprises the following steps that the number of rows of 1-w pixels in a pixel matrix of a scrap paper sheet is 1 in gray level binary value, the number of rows of 1-w pixels in gray level binary value is 1 in gray level binary value is used as the upper edge distance of an original file, and w is a positive integer larger than 1 and smaller than m;
step S6', determining other paper scraps of which the line number from the 1 st line pixel to the w th line pixel in the pixel matrix is equal to the upper margin of the original file and the gray binarization values of all 1, and obtaining C1，2Shredding to C1，g-1Scrap paper, thereby determining the position h1The shredded paper of (1);
step S7' obtaining signals at l1C in (1)f，1Shredded paper and a sheet located ingC in (1)f，gPixel matrix of a chip sheet, record Cf，1Paper scraps and Cf，gThe method comprises the following steps that the number of rows of m-th row pixels to m-t-th row pixels in a pixel matrix of a scrap paper sheet is 1, the number of rows of m-th row pixels to m-t row pixels with the total gray level binarization values of 1 is used as the lower margin of an original file, and t is a positive integer smaller than m;
step S8', determining other paper scraps of which the line number from the m-th line pixel to the m-t-th line pixel in the pixel matrix is equal to the upper margin of the original file and the gray binary value of which is 1, and obtaining Cf，2Shredding to Cf，g-1Scrap paper, thereby determining the position hfThe shredded paper of (1);
step S9' according to the position l1、lg、h1、hfThe paper scraps of the paper are spliced together to restore the original document.
The original document is a printed text document on the same page, and the paper scraps are paper scraps of the original document transversely cut in the longitudinal direction.
In the step S3', it is determined that the position is l1The specific process of the paper scraps in the process comprises the following steps: will EnH of the Q-th scrap of one scrap1 QH with Q' th scrap piecen Q’The absolute distance matrix of the first and second paper scraps is used for judging that the Q-th paper scrap and the Q' -th paper scrap are in the position of l1In the position where the Q-th scrap piece is located at l when the absolute distance matrix is 01C in (1)1，1(ii) a And again according to h of the Q-th scrap piecen QAnd EnArranging the remaining paper scraps in the absolute distance matrix between the pixel gray value vectors of the 1 st column of the remaining paper scraps in the paper scraps in sequence according to the relationship from small to large of the absolute distance matrix1C of (A)2，1To Cf，1Is prepared by1Each of which is rotated 90 deg. counterclockwise to be positioned at l1The shredded paper of (1).
The specific process of the step S9' is as follows:
step S91', setting the paper scrap with undetermined position as the paper scrap in idle state, and calculating the upper boundary and C of the paper scrap in idle statef’-1，g’Absolute distance of lower boundary of paper scrap and left boundary and C of paper scrap in idle statef’，g’-1Absolute distance of the right boundary of the scrap piece;
step S92' passing through the matching degree formula
Counting idle scrap and Cf’，g’Obtaining P paper scraps of which the matching degrees are sequentially reduced from the highest to the lowest by the matching degrees of the paper scraps;
wherein, Vf’，g’Degree of match, D ', of sheets of paper in idle state to sheets of paper'1Absolute distance, D ', of the upper boundary of the sheet of paper in an idle state from the lower boundary of the sheet of paper'2Setting f 'to be more than or equal to 2 and less than or equal to f-1, g' to be more than or equal to 2 and less than or equal to g-1, and P is a positive integer more than or equal to 2;
step S93', putting P paper scraps into the C paper scraps in sequence from high to low according to the matching degreef’，g’Position, by Cf’-1，g’Paper scraps and Cf’，g’-1The character information of the paper scraps and the correctly matched paper scraps are judged by the vision of human eyes, thereby determining Cf’，g’A shredded paper sheet;
step S94 ', repeat steps S91 ' through S93 ' until determining C2，2Shredding to Cf-1，g-1Scrap the paper, thereby restoring the entire original document.
And P is 10.
In step S1', the plurality of shredded paper pieces are converted into a plurality of images in a computer format by a scanner, and the gray value of each pixel in the plurality of images is read by computer software.
The invention has the beneficial effects that: the original file restoration method of the paper scraps comprises the steps of determining the paper scraps located on the leftmost side of an original file, sequentially arranging other paper scraps on the right side of the paper scraps located on the leftmost side according to the absolute distance matrix relation between the other paper scraps and the paper scraps located on the leftmost side of the original file from small to large of the absolute distance matrix, and accordingly restoring the original file, or determining the left boundary and the right boundary of a restoration matrix located on the original file, determining the upper boundary and the lower boundary of the restoration matrix according to the left boundary and the right boundary of the restoration matrix, combining human vision with a computer, and splicing the paper scraps in an idle state by using a human-computer interaction interface, so that the whole original file is restored, a large number of human resources are saved, the working efficiency is improved, and the restoration accuracy is high.
Drawings
For a better understanding of the nature and technical aspects of the present invention, reference should be made to the following detailed description of the invention, taken in conjunction with the accompanying drawings, which are provided for purposes of illustration and description and are not intended to limit the invention.
In the drawings, there is shown in the drawings,
FIG. 1 is a flowchart of a first embodiment of a method for restoring an original document of a scrap paper sheet according to the present invention;
fig. 2 is a flowchart of a method for restoring an original document of a scrap paper according to a second embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention, the following detailed description is given with reference to the preferred embodiments of the present invention and the accompanying drawings.
Referring to fig. 1, the present invention provides a first embodiment of a method for restoring an original document of a paper scrap, including the following steps:
step S1, obtaining a plurality of paper fragments corresponding to the original document and a plurality of images corresponding to the paper fragments, reading a gray value of each pixel in the plurality of images, and obtaining a pixel matrix corresponding to each paper fragment: a. thei＝(a1，a2，Λ，an) Wherein A isiPixel matrix corresponding to the ith scrap piece, a1For the 1 st column pixel gray value vector, a, of the ith scrapnSetting i as a positive integer and n as a positive integer larger than 2 for the pixel gray value vector of the nth column of the ith paper scrap;
step S2, processing each row of pixel gray value vectors in the pixel matrix corresponding to each paper scrap according to a binarization formula to obtain a binarization value of each component value in each row of pixel gray value vectors of each paper scrap;
the binarization formula is as follows:
wherein R isi n(k) Is the binary value of the kth component value in the n-th column pixel gray value vector of the ith wastepaper sheet, an(k) Setting k as a natural number more than 2 for the kth component value in the n-th column of pixel gray value vector of the ith scrap paper sheet;
step S3, determining the position of the ith paper scrap at the original file according to the binary value of each component value in the first to Z-th rows of pixel gray-scale value vectors of the ith paper scrap, determining that the ith paper scrap is located at the leftmost position of the original file when the binary value of each component value in the first to Z-th rows of pixel gray-scale value vectors of the ith paper scrap is 1, and setting Z to be a natural number greater than 1 (generally, the original file has a specific margin, and when the binary value of each component value in the preceding Z-th row of pixel gray-scale value vectors of the ith paper scrap is 1, we can determine that the paper scrap is the leftmost position of the file, the size of the Z value is not fixed, and can select according to the margin of the actual original file), or,
judging the positions of the ith paper scrap and the jth paper scrap on the original file according to an absolute distance matrix between the binary value of each component value in the 1 st column of pixel gray value vectors of the ith paper scrap and the binary value of each component value in the nth column of pixel gray value vectors of the jth paper scrap, wherein when the absolute distance matrix is 0, the ith paper scrap is positioned at the leftmost side of the original file, the jth paper scrap is positioned at the rightmost side of the original file, and j is set as a positive integer;
and step S4, according to an absolute distance matrix between the binary value of each component value in the nth column of pixel gray value vectors of the ith paper scrap and the binary value of each component value in the 1 st column of pixel gray value vectors of other paper scrap, arranging other paper scrap on the right side of the ith paper scrap in sequence according to the relationship from small to large of the absolute distance matrix, and thus restoring the original file.
Specifically, the original document is a printed text document on the same page, and the paper scraps are longitudinally cut from the original document.
Specifically, in step S1, the 1 st column of pixel gray-scale value vectors refer to gray-scale values of the 1 st column of pixels in the 1 st row arranged in sequence from the gray-scale value of the 1 st column of pixels in the 1 st row of pixels.
Specifically, in step S1, the scanner converts the multiple pieces of paper into multiple images in a computer format, such as a bmp format, in which the images are arranged in a pixel array, so that the computer software (such as Matlab software) can be used to read the grayscale value of each pixel in the multiple images.
When Matlab software is used for reading the gray value of each pixel in a plurality of images, the gray value of a part representing a white bottom can be seen from the gray value of each pixel, and is 255, and the gray value occupies most of the inside of a paper scrap and the boundary of the paper scrap; the gray value of the black part representing the text content is smaller than 255, the closer to black, the smaller the gray value of the black part is, the gray value of pure black is 0, and the distribution of the black part in the picture is sparse.
Specifically, a binary value of 1 indicates white, and a binary value of 0 indicates black.
Specifically, the 1 st column of pixel gradation value vector of the ith paper scrap is equivalent to the left boundary vector of the ith paper scrap, and the n th column of pixel gradation value vector of the ith paper scrap is equivalent to the right boundary vector of the ith paper scrap. In step S3, the absolute distance matrix between the binarized value of each component value in the 1 st column of pixel grayscale value vector of the ith paper scrap and the binarized value of each component value in the n th column of pixel grayscale value vector of the jth paper scrap is: dij＝∑|Ri 1(k)-Rj n(k) L, wherein DijIs an absolute distance matrix, R, between the 1 st column pixel gray value vector of the ith paper scrap and the n th column pixel gray value vector of the jth paper scrapi 1(k) Is the binary value, R, of the kth component value in the 1 st column pixel gray value vector of the ith scrap sheetj n(k) A binary value of a k component value in a pixel gray value vector of an nth column of a jth paper scrap sheet; (that is, an element in the absolute distance matrix is 0 and is only one 0 in the matrix, which means that the left boundary of the ith scrap is completely matched with the right boundary of the jth scrap because all pixels on the left boundary of the ith scrap and the right boundary of the jth scrap are white, and because there is a gap between characters, all pixels cannot be black, i.e., the 1 st column of pixels of the ith scrap is the leftmost side of the original document, the nth column of pixels of the jth scrap is the rightmost side of the original document, the ith scrap is located at the leftmost side of the original document, and the jth scrap is located at the rightmost side of the original document).
In particular, DijThe larger the value of (a) indicates the absolute distance between the left boundary of the ith scrap and the right boundary of the jth scrapThe larger the distance, the lower their degree of matching. All the scraps of paper can be spliced in a mode of arranging from left to right according to the distance between the left and right boundaries of different scraps of paper. It should be noted that the distance between the pieces of paper is related to the order in which they are arranged, in general DijIs not equal to DjiI.e. the resulting distance matrix is not an upper or lower triangular matrix; when i ═ j, DijInfinite, which means that the left and right boundaries of the scrap piece itself do not match.
In step S4, the absolute distance matrix between the binarized value of each component value in the n-th column pixel grayscale value vector of the i-th scrap piece and the binarized value of each component value in the 1-th column pixel grayscale value vector of the other scrap piece is: dio＝∑|Ri n(k)-Ro 1(k) L, wherein DioIs an absolute distance matrix, R, between the gray value vector of the nth column pixel of the ith paper scrap and the gray value vector of the 1 st column pixel of other paper scrapi n(k) Is the binary value, R, of the kth component value in the n-th column pixel gray-scale value vector of the ith scrap sheeto 1(k) The binary value of the kth component value in the 1 st column pixel gray value vector of the other paper scrap is obtained.
It should be noted that, for longitudinally cut paper scraps, the invention firstly determines the paper scraps located at the leftmost side of the original document, and then sequentially arranges other paper scraps at the right side of the leftmost paper scraps according to the absolute distance matrix relationship between the other paper scraps and the leftmost paper scraps of the original document and the relationship from small to large of the absolute distance matrix, thereby restoring the original document, saving a large amount of human resources, improving the working efficiency and having high restoration accuracy.
Referring to fig. 2, the present invention further provides a second embodiment of a method for restoring an original document of a shredded paper, including the following steps:
step S1', obtaining a plurality of paper fragments corresponding to the original document and a plurality of images corresponding to the plurality of paper fragments, reading a gray value of each pixel in the plurality of images, and obtaining a pixel matrix corresponding to each paper fragment:
wherein, BkPixel matrix for the kth scrap piece, bm，n kIs the gradation value of the pixel of the mth row and nth column of the kth chip sheet, ln kFor the pixel gray value vector of the nth column of the kth chip sheet, hm kSetting k as a positive integer and setting m and n as positive integers larger than 2 for the pixel gray value vector of the mth row of the kth paper scrap;
step S2', the gray value of each pixel in the pixel matrix corresponding to each shredded paper is processed according to the binarization formula, and the gray binarization value of each pixel of each shredded paper is obtained;
the binarization formula is as follows:
wherein x is 1, 2, ·, m; y is 1, 2,. cndot.n; b'x，y kA gray level binary value of the pixel of the x row and the y column of the k scrap paper sheet;
step S3', set the restoration matrix of the original file as
Wherein, Cf×gTo restore the matrix, Cf，gFor scraps of paper located in the f-th row and g-th column,/gFor scraps of paper in the g-th column, hfIs a scrap piece located in row f; (generally speaking, the smaller the amount of information on the fragments, the greater the probability of splicing errors or even non-splicing, so the feature information of the paper scraps should be applied as comprehensively as possiblef×gWe refer to the restoration matrix, i.e. each element in the restoration matrix represents a piece of shredded paper, and each element can be referred to as a unit); record l1 kThe number of the pieces of paper which are all 1 is E1When E is1>f is then at E1Record l in one scrap piece2 kThe number of the pieces of paper which are all 1 isE2When E is2>f is then at E2Record l in one scrap piece3 kThe number of all 1 scraps was E3, up to EnIf f, then determine ln kAll 1EnThe individual scraps of paper are positioned in the first column of the restoration matrix, thus being positioned in l1The shredded paper of (1);
step S4', record ln kThe number of the pieces of paper which are all 1 is En', when En’>f is then at En' record in one scrap piecen-1 kThe number of the pieces of paper which are all 1 is En-1', when En-1’>f is then at En-1' record in one scrap piecen-2 kThe number of the pieces of paper which are all 1 is En-2', up to En-pIf f, then determine ln-p kAll 1En-p' one scrap piece is located in the last column of the recovery matrix, thus being located exactly ingSetting p as a positive integer smaller than n;
step S5' obtaining signals at l1C in (1)1，1Shredded paper and a sheet located ingC in (1)1，gPixel matrix of a chip sheet, record C1，1Paper scraps and C1，gThe method comprises the following steps that the number of rows of 1-w pixels in a pixel matrix of a scrap paper sheet is 1 in gray level binary value, the number of rows of 1-w pixels in gray level binary value is 1 in gray level binary value is used as the upper edge distance of an original file, and w is a positive integer larger than 1 and smaller than m;
step S6', determining other paper scraps of which the line number from the 1 st line pixel to the w th line pixel in the pixel matrix is equal to the upper margin of the original file and the gray binarization values of all 1, and obtaining C1，2Shredding to C1，g-1Scrap paper, thereby determining the position h1The shredded paper of (1);
step S7' obtaining signals at l1C in (1)f，1Shredded paper and a sheet located ingC in (1)f，gPixel matrix of a chip sheet, record Cf，1Paper scraps and Cf，gM-th to m-th rows of pixels in a pixel matrix of a scrap sheet-the number of rows where the binary values of the gray levels of the pixels in t rows are all 1, the number of rows where the binary values of the gray levels of the pixels in the m-th row to the m-t row are all 1 is taken as the bottom margin of the original file, and t is a positive integer smaller than m;
step S8', determining other paper scraps of which the line number from the m-th line pixel to the m-t-th line pixel in the pixel matrix is equal to the upper margin of the original file and the gray binary value of which is 1, and obtaining Cf，2Shredding to Cf，g-1Scrap paper, thereby determining the position hfThe shredded paper of (1);
step S9' according to the position l1、lg、h1、hfThe paper scraps of the paper are spliced together to restore the original document.
Specifically, the original document is a printed text document on the same page, the paper scraps are paper scraps of the original document transversely and longitudinally cut, at this time, the restoration work cannot be completed only by means of information on the left and right boundaries of each paper scrap, and all pixel point information is needed.
Specifically, in the step S1', the pixel gray-scale value vectors of the nth column refer to the gray-scale values of the pixels in the 1 st row of the nth column and the pixels in the last 1 st row of the nth column which are sequentially arranged; the pixel gray value vector of the mth row refers to the gray value of the 1 st column of pixels of the mth row-column arranged in sequence from the gray value of the 1 st column of pixels of the mth column-row to the gray value of the last 1 st column of pixels of the mth column-column.
Specifically, in step S1', the scanner converts the multiple paper scraps into multiple images in a computer format, such as a bmp format, in which the images are arranged in a pixel array, so that the computer software (such as Matlab software) can be used to read the grayscale value of each pixel in the multiple images.
When Matlab software is used for reading the gray value of each pixel in a plurality of images, the gray value of a part representing a white bottom can be seen from the gray value of each pixel, and is 255, and the gray value occupies most of the inside of a paper scrap and the boundary of the paper scrap; the gray value of the black part representing the text content is smaller than 255, the closer to black, the smaller the gray value of the black part is, the gray value of pure black is 0, and the distribution of the black part in the picture is sparse.
Specifically, a binary value of 1 indicates white, and a binary value of 0 indicates black.
Specifically, in the step S3', the position/may be determined by the first embodiment of the method for restoring the original document of the scrap paper, i.e., by the method for restoring the scrap paper by slitting1The shredded paper of (1). The method specifically comprises the following steps: will EnH of the Q-th scrap of one scrap1 QH with Q' th scrap piecen Q’The absolute distance matrix of the first and second paper scraps is used for judging that the Q-th paper scrap and the Q' -th paper scrap are in the position of l1In the position where the Q-th scrap piece is located at l when the absolute distance matrix is 01C in (1)1，1(ii) a And again according to h of the Q-th scrap piecen QAnd EnArranging the remaining paper scraps in the absolute distance matrix between the pixel gray value vectors of the 1 st column of the remaining paper scraps in the paper scraps in sequence according to the relationship from small to large of the absolute distance matrix1C of (A)2，1To Cf，1Is prepared by1Each of which is rotated 90 deg. counterclockwise to be positioned at l1The shredded paper of (1).
Further, in the step S4', it is also determined that the position l is located by a method of restoring the longitudinally cut shredded papergThe scrap in (h) can also be determined by the scrap recovery method of the slit scrap in (S6')1The scrap in (h) can also be determined by the scrap recovery method of the slit scrap in (S8')fThe paper scraps in (1) are not described in detail herein.
Specifically, the specific process of step S9' is as follows:
step S91', setting the paper scrap with undetermined position as the paper scrap in idle state, and calculating the upper boundary and C of the paper scrap in idle statef’-1，g’Absolute distance of lower boundary of paper scrap and left boundary and C of paper scrap in idle statef’，g’-1Absolute distance of the right boundary of the scrap piece;
step S92' passing through the matching degree formula
Counting idle scrap and Cf’，g’Obtaining P paper scraps of which the matching degrees are sequentially reduced from the highest to the lowest by the matching degrees of the paper scraps;
wherein, Vf’，g’Degree of match, D ', of sheets of paper in idle state to sheets of paper'1Absolute distance, D ', of the upper boundary of the sheet of paper in an idle state from the lower boundary of the sheet of paper'2Setting f 'to be more than or equal to 2 and less than or equal to f-1, g' to be more than or equal to 2 and less than or equal to g-1, and P is a positive integer more than or equal to 2;
step S93', putting P paper scraps into the C paper scraps in sequence from high to low according to the matching degreef’，g’Position, by Cf’-1，g’Paper scraps and Cf’，g’-1The character information of the paper scraps and the correctly matched paper scraps are judged by the vision of human eyes, thereby determining Cf’，g’A shredded paper sheet;
step S94 ', repeat steps S91 ' through S93 ' until determining C2，2Shredding to Cf-1，g-1Scrap the paper, thereby restoring the entire original document.
Specifically, in order to ensure the accuracy of the judgment, P is preferably 10.
It should be noted that, for the longitudinally and transversely cut paper scraps, the invention firstly determines the paper scraps positioned at the left boundary and the right boundary of the original file, namely, determines the left boundary and the right boundary of the restoration matrix, then determines the upper boundary and the lower boundary of the restoration matrix according to the left boundary and the right boundary of the restoration matrix, finally combines the human vision with the computer, and splices the paper scraps in an idle state by using a human-computer interaction interface, thereby restoring the whole original file, saving a large amount of human resources, improving the working efficiency and having high restoration accuracy.
In summary, the original file restoration method of the paper scraps of the present invention restores the original file by determining the leftmost paper scraps of the original file, and then sequentially arranging the other paper scraps on the right side of the leftmost paper scraps according to the absolute distance matrix relationship between the other paper scraps and the leftmost paper scraps of the original file from small to large of the absolute distance matrix, or determining the left boundary and the right boundary of the restoration matrix of the original file, then determining the upper boundary and the lower boundary of the restoration matrix according to the left boundary and the right boundary of the restoration matrix, and finally combining the human vision with the computer, and splicing the paper scraps in an idle state by using the human-computer interaction interface, thereby restoring the entire original file, saving a large amount of human resources, improving the work efficiency, and having high restoration accuracy.
As described above, it will be apparent to those skilled in the art that other various changes and modifications may be made based on the technical solution and concept of the present invention, and all such changes and modifications are intended to fall within the scope of the appended claims.
Claims (10)
1. A method for restoring original documents of paper scraps is characterized by comprising the following steps:
step S1, obtaining a plurality of paper fragments corresponding to the original document and a plurality of images corresponding to the paper fragments, reading a gray value of each pixel in the plurality of images, and obtaining a pixel matrix corresponding to each paper fragment: a. thei＝(a1，a2，…，an) Wherein A isiPixel matrix corresponding to the ith scrap piece, a1For the 1 st column pixel gray value vector, a, of the ith scrapnSetting i as a positive integer and n as a positive integer larger than 2 for the pixel gray value vector of the nth column of the ith paper scrap;
step S2, processing each row of pixel gray value vectors in the pixel matrix corresponding to each paper scrap according to a binarization formula to obtain a binarization value of each component value in each row of pixel gray value vectors of each paper scrap;
the binarization formula is as follows:
wherein R isi n(k) For the nth row pixel ash of the ith paper scrapBinary value of the kth component value in the value vector, an(k) Setting k as a natural number more than 2 for the kth component value in the n-th column of pixel gray value vector of the ith scrap paper sheet;
step S3, judging the positions of the ith paper scrap and the jth paper scrap on the original file according to an absolute distance matrix between the binary value of each component value in the 1 st column of pixel gray value vectors of the ith paper scrap and the binary value of each component value in the nth column of pixel gray value vectors of the jth paper scrap, wherein when the absolute distance matrix is 0, the ith paper scrap is positioned at the leftmost side of the original file, the jth paper scrap is positioned at the rightmost side of the original file, and j is set as a positive integer;
and step S4, according to an absolute distance matrix between the binary value of each component value in the nth column of pixel gray value vectors of the ith paper scrap and the binary value of each component value in the 1 st column of pixel gray value vectors of other paper scrap, arranging other paper scrap on the right side of the ith paper scrap in sequence according to the relationship from small to large of the absolute distance matrix, and thus restoring the original file.
2. The method for restoring an original document of a paper scrap according to claim 1 wherein said original document is a printed text document of the same page and said paper scrap is a paper scrap cut longitudinally from said original document.
3. The method for restoring an original document in a paper scrap as claimed in claim 1, wherein in said step S3, the absolute distance matrix between the binarized value of each component value in the 1 st column pixel gray scale value vector of the ith paper scrap and the binarized value of each component value in the n th column pixel gray scale value vector of the jth paper scrap is: dij＝∑|Ri 1(k)-Rj n(k) L, wherein DijIs an absolute distance matrix, R, between the 1 st column pixel gray value vector of the ith paper scrap and the n th column pixel gray value vector of the jth paper scrapi 1(k) Is the binary value, R, of the kth component value in the 1 st column pixel gray value vector of the ith scrap sheetj n(k) A binarized value of a k-th component value in the n-th column pixel gradation value vector for the jth chip sheet.
4. The method for restoring an original document in a paper scrap as claimed in claim 3, wherein in said step S4, the absolute distance matrix between the binarized value of each component value in the n-th column pixel gray scale value vector of the i-th paper scrap and the binarized value of each component value in the 1-th column pixel gray scale value vector of the other paper scrap is: dio＝∑|Ri n(k)-Ro 1(k) L, wherein DioIs an absolute distance matrix, R, between the gray value vector of the nth column pixel of the ith paper scrap and the gray value vector of the 1 st column pixel of other paper scrapi n(k) Is the binary value, R, of the kth component value in the n-th column pixel gray-scale value vector of the ith scrap sheeto 1(k) The binary value of the kth component value in the 1 st column pixel gray value vector of the other paper scrap is obtained.
5. A method for restoring original documents of paper scraps is characterized by comprising the following steps:
step S1', obtaining a plurality of paper fragments corresponding to the original document and a plurality of images corresponding to the paper fragments, reading a gray value of each pixel in the plurality of images, and obtaining a pixel matrix corresponding to each paper fragment:
wherein, BkPixel matrix for the kth scrap piece, bm，n kIs the gradation value of the pixel of the mth row and nth column of the kth chip sheet, ln kFor the pixel gray value vector of the nth column of the kth chip sheet, hm kSetting k as a positive integer and setting m and n as positive integers larger than 2 for the pixel gray value vector of the mth row of the kth paper scrap;
step S2', the gray value of each pixel in the pixel matrix corresponding to each shredded paper is processed according to the binarization formula, and the gray binarization value of each pixel of each shredded paper is obtained;
the binarization formula is as follows:
wherein x is 1, 2, ·, m; y is 1, 2,. cndot.n; b'x，y kA gray level binary value of the pixel of the x row and the y column of the k scrap paper sheet;
step S3', set the restoration matrix of the original file as
Wherein, Cf×gTo restore the matrix, Cf，gFor scraps of paper located in the f-th row and g-th column,/gFor scraps of paper in the g-th column, hfIs a scrap piece located in row f; record l1 kThe number of the pieces of paper which are all 1 is E1When E is1>f is then at E1Record l in one scrap piece2 kThe number of the pieces of paper which are all 1 is E2When E is2>f is then at E2Record l in one scrap piece3 kThe number of all 1 scraps was E3, up to EnIf f, then determine ln kAll 1EnThe individual scraps of paper are positioned in the first column of the restoration matrix, thus being positioned in l1The shredded paper of (1);
step S4', record ln kThe number of the pieces of paper which are all 1 is En', when En’>f is then at En' record in one scrap piecen-1 kThe number of the pieces of paper which are all 1 is En-1', when En-1’>f is then at En-1' record in one scrap piecen-2 kThe number of the pieces of paper which are all 1 is En-2', up to En-pIf f, then determine ln-p kAll 1En-p' one scrap piece is located in the last column of the recovery matrix, thus being located exactly ingSetting p as a positive integer smaller than n;
step S5' obtaining signals at l1C in (1)1，1Shredded paper and a sheet located ingC in (1)1，gPixel matrix of a chip sheet, record C1，1Paper scraps and C1，gThe method comprises the following steps that the number of rows of 1-w pixels in a pixel matrix of a scrap paper sheet is 1 in gray level binary value, the number of rows of 1-w pixels in gray level binary value is 1 in gray level binary value is used as the upper edge distance of an original file, and w is a positive integer larger than 1 and smaller than m;
step S6', determining other paper scraps of which the line number from the 1 st line pixel to the w th line pixel in the pixel matrix is equal to the upper margin of the original file and the gray binarization values of all 1, and obtaining C1，2Shredding to C1，g-1Scrap paper, thereby determining the position h1The shredded paper of (1);
step S7' obtaining signals at l1C in (1)f，1Shredded paper and a sheet located ingC in (1)f，gPixel matrix of a chip sheet, record Cf，1Paper scraps and Cf，gThe method comprises the following steps that the number of rows of m-th row pixels to m-t-th row pixels in a pixel matrix of a scrap paper sheet is 1, the number of rows of m-th row pixels to m-t row pixels with the total gray level binarization values of 1 is used as the lower margin of an original file, and t is a positive integer smaller than m;
step S8', determining other paper scraps of which the line number from the m-th line pixel to the m-t-th line pixel in the pixel matrix is equal to the upper margin of the original file and the gray binary value of which is 1, and obtaining Cf，2Shredding to Cf，g-1Scrap paper, thereby determining the position hfThe shredded paper of (1);
step S9' according to the position l1、lg、h1、hfThe paper scraps of the paper are spliced together to restore the original document.
6. The method for restoring an original document made of paper scraps as set forth in claim 5, wherein said original document is a printed text document of the same page, and said paper scraps are paper scraps of the original document transversely cut in the longitudinal direction.
7. Such asThe method for restoring original documents of paper scraps as claimed in claim 5, wherein in said step S3', it is determined that the position is in l1The specific process of the paper scraps in the process comprises the following steps: will EnH of the Q-th scrap of one scrap1 QH with Q' th scrap piecen Q’The absolute distance matrix of the first and second paper scraps is used for judging that the Q-th paper scrap and the Q' -th paper scrap are in the position of l1In the position where the Q-th scrap piece is located at l when the absolute distance matrix is 01C in (1)1，1(ii) a And again according to h of the Q-th scrap piecen QAnd EnArranging the remaining paper scraps in the absolute distance matrix between the pixel gray value vectors of the 1 st column of the remaining paper scraps in the paper scraps in sequence according to the relationship from small to large of the absolute distance matrix1C of (A)2，1To Cf，1Is prepared by1Each of which is rotated 90 deg. counterclockwise to be positioned at l1The shredded paper of (1).
8. The method for restoring an original document in a paper scrap according to claim 5, wherein the step S9' is embodied by the process of:
step S91', setting the paper scrap with undetermined position as the paper scrap in idle state, and calculating the upper boundary and C of the paper scrap in idle statef’-1，g’Absolute distance of lower boundary of paper scrap and left boundary of paper scrap in idle state
Absolute distance of the right boundary of the scrap piece;
step S92' passing through the matching degree formula
Counting idle scrap and Cf’，g’Obtaining P paper scraps of which the matching degrees are sequentially reduced from the highest to the lowest by the matching degrees of the paper scraps;
wherein, Vf’，g’Degree of match, D ', of sheets of paper in idle state to sheets of paper'1Is in idle stateAbsolute distance, D ', of upper boundary of the as-formed paper shredder sheet from lower boundary of the paper shredder sheet'2Setting f 'to be more than or equal to 2 and less than or equal to f-1, g' to be more than or equal to 2 and less than or equal to g-1, and P is a positive integer more than or equal to 2;
step S93', putting P paper scraps into the C paper scraps in sequence from high to low according to the matching degreef’，g’Position, by Cf’-1，g’Paper scraps and
the character information of the paper scraps and the correctly matched paper scraps are judged by the vision of human eyes, thereby determining Cf’，g’A shredded paper sheet;
step S94 ', repeat steps S91 ' through S93 ' until determining C2，2Shredding to Cf-1，g-1Scrap the paper, thereby restoring the entire original document.
9. The method for restoring an original document from a paper scrap as set forth in claim 8, wherein P is 10.
10. The method for restoring an original document in a paper scrap according to claim 5 wherein in step S1', the plurality of paper scrap pieces are converted into a plurality of images in a computer format by a scanner, and the gray value of each pixel in the plurality of images is read by computer software.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811544621.0A CN109584163B (en) | 2018-12-17 | 2018-12-17 | Method for restoring original file of paper scrap |
| PCT/CN2018/125072 WO2020124676A1 (en) | 2018-12-17 | 2018-12-28 | Method for restoring original file of shredded paper |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811544621.0A CN109584163B (en) | 2018-12-17 | 2018-12-17 | Method for restoring original file of paper scrap |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109584163A CN109584163A (en) | 2019-04-05 |
| CN109584163B true CN109584163B (en) | 2020-12-08 |
Family
ID=65930554
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811544621.0A Active CN109584163B (en) | 2018-12-17 | 2018-12-17 | Method for restoring original file of paper scrap |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109584163B (en) |
| WO (1) | WO2020124676A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110246098B (en) * | 2019-05-31 | 2021-07-27 | 暨南大学 | Fragment recovery method |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE19803127A1 (en) * | 1998-01-28 | 1999-08-05 | Foerderung Angewandter Informa | Document reassembly method for shredded document sheet |
| TWI394098B (en) * | 2009-06-03 | 2013-04-21 | Nat Univ Chung Cheng | Shredding Method Based on File Image Texture Feature |
| CN102890818B (en) * | 2012-08-06 | 2015-06-17 | 厦门市美亚柏科信息股份有限公司 | Method and device for reassembling joint photographic experts group (JPG) picture fragments based on thumbnail |
| CN103020929B (en) * | 2012-11-26 | 2015-08-26 | 西安理工大学 | The broken document recovery method of shredder based on character features |
| CN103700081B (en) * | 2013-12-17 | 2016-08-17 | 河海大学 | A kind of shredder crushes the restoration methods of English document |
| CN103679678B (en) * | 2013-12-18 | 2016-11-23 | 山东大学 | A kind of semi-automatic splicing restored method of rectangle character features a scrap of paper |
| CN103996180B (en) * | 2014-05-05 | 2016-09-07 | 河海大学 | Shredder based on English words feature crushes document restored method |
| CN104182966B (en) * | 2014-07-16 | 2017-03-29 | 江苏大学 | A kind of regular shredded paper method for automatically split-jointing |
| CN104143095B (en) * | 2014-07-16 | 2017-10-03 | 暨南大学 | Reconstruction of fragments method based on genetic algorithm and character recognition technologies |
| CN104657964B (en) * | 2015-03-09 | 2017-10-31 | 重庆邮电大学 | A kind of fragment image computer method for automatically split-jointing |
| EP3093851B1 (en) * | 2015-05-11 | 2017-11-08 | Micro Systemation AB | Method and device for use when reassembling a fragmented jpeg image |
| CN106952230B (en) * | 2017-03-19 | 2021-02-02 | 北京工业大学 | Transverse and longitudinal shredded slice restoration method based on clustering and ant colony algorithm |
| CN108510442B (en) * | 2018-03-23 | 2021-12-31 | 中南大学 | Single-side paper scrap splicing and restoring method based on absolute value distance optimization |
| CN108921793A (en) * | 2018-07-15 | 2018-11-30 | 江西理工大学 | A scrap of paper based on fuzzy C-means clustering splices restored method |
-
2018
- 2018-12-17 CN CN201811544621.0A patent/CN109584163B/en active Active
- 2018-12-28 WO PCT/CN2018/125072 patent/WO2020124676A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020124676A1 (en) | 2020-06-25 |
| CN109584163A (en) | 2019-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10970535B2 (en) | System and method for extracting tabular data from electronic document | |
| CN111080660B (en) | Image segmentation method, device, terminal equipment and storage medium | |
| CN103700081B (en) | A kind of shredder crushes the restoration methods of English document | |
| EP2178028A2 (en) | Representing documents with runlength histograms | |
| Lin et al. | Reconstruction of shredded document based on image feature matching | |
| CN112926565B (en) | Picture text recognition method, system, equipment and storage medium | |
| CN114972922B (en) | Coal gangue sorting and identifying method, device and equipment based on machine learning | |
| CN104182722A (en) | Text detection method and device and text information extraction method and system | |
| CN104182966B (en) | A kind of regular shredded paper method for automatically split-jointing | |
| CN109584163B (en) | Method for restoring original file of paper scrap | |
| DE202022104365U1 (en) | A robust color image hashing system for image authentication | |
| CN114528950A (en) | Destroying method and system for identifying type of confidential medium based on three-dimensional point cloud | |
| US10095677B1 (en) | Detection of layouts in electronic documents | |
| CN102737240A (en) | Method of analyzing digital document images | |
| CN111369489B (en) | Image identification method and device and terminal equipment | |
| CN108510442B (en) | Single-side paper scrap splicing and restoring method based on absolute value distance optimization | |
| CN114495139A (en) | Operation duplicate checking system and method based on image | |
| Padma et al. | I DENTIFICATION OF T ELUGU, D EVANAGARI AND E NGLISH S CRIPTS U SING D ISCRIMINATING | |
| EP1299857B1 (en) | Image processing method and system | |
| US9053359B2 (en) | Method and system for document authentication using Krawtchouk decomposition of image patches for image comparison | |
| De Smet et al. | Semiautomatic reconstruction of strip-shredded documents | |
| CN107180412B (en) | Horizontal and vertical shredded paper sheet reconstruction method based on horizontal projection and K-means clustering | |
| CN106952230B (en) | Transverse and longitudinal shredded slice restoration method based on clustering and ant colony algorithm | |
| Madain | Clustering paper shreds of different sizes | |
| Rao et al. | Script identification of telugu, english and hindi document image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |