CN103085471A - Positioning method for continuous acquisition of presswork images - Google Patents
Positioning method for continuous acquisition of presswork images Download PDFInfo
- Publication number
- CN103085471A CN103085471A CN2013100117063A CN201310011706A CN103085471A CN 103085471 A CN103085471 A CN 103085471A CN 2013100117063 A CN2013100117063 A CN 2013100117063A CN 201310011706 A CN201310011706 A CN 201310011706A CN 103085471 A CN103085471 A CN 103085471A
- Authority
- CN
- China
- Prior art keywords
- image
- row
- certain
- printed matter
- images
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention belongs to the field of presswork full picture quality detection and quality control, and provides a positioning method for a continuous acquisition of presswork images. The positioning method for the continuous acquisition of presswork images comprises the steps of obtaining single channel images with the richest information in photoelectric coupler collecting images, taking the absolute value of difference between all pixel values of the single channel images and the pixel values of papers of printed materials by the single channel, making cumulative statistics to the absolute value of difference according to lines and rows, judging four boundaries of images of adjacent two line and two row statistics along an upper direction, a lower direction, a left direction, and a right direction according to boundary judging sequence and judging condition, and thereby finishing positioning to printing images. In addition, auxiliary tags of packaging pressworks are positioned through strip shapes, and channels with rich information and calculation of absolute value of difference are not needed. According to the positioning method for the continuous acquisition of presswork images, positioning of continuous acquisition of presswork images of webs, single sheet-fed printing, and package printing is achieved, practicality is strong, affect on image positioning from density of printing images and package printing and the like is avoided, and positioning accuracy is high.
Description
Technical field
The invention belongs to the quality testing of printed matter full frame and control field, relate in particular to a kind of localization method to continuous acquisition printed matter image.
Background technology
Current printing industry has entered full-automation, high speed and digitized epoch fully, only detect and control the methods such as printing quality with the experience of printing worker's long-term accumulation, printed matter sampling Detection in printing process and traditional printing control strip, no longer can satisfy the requirement that printing machine performance, printing material are adjusted in real time and print quality is judged fast, these methods have been no longer to guarantee that printed matter is high-quality, printing process efficient and printing trader's best means cheaply.At present, along with improving constantly of computer application the reach of science and printing industry automatization level, the advantages such as NI Vision Builder for Automated Inspection is large with the density of himself obtaining information, speed is fast, full-automatic, objectivity and accuracy, can satisfy the requirement of real-time that current print quality detects and control, comprehensive and accuracy fully, therefore the quality testing of printed matter full frame and the control technology based on machine vision becomes the current research focus.
At present, when utilizing photoelectrical coupler that image is gathered based on the printed matter full frame quality detecting system of machine vision, can because of factors such as collecting device mechanical reason and external environmental interference, cause same photoelectrical coupler spatially to there are differences at the image of different time collection.Meanwhile, all adopt testing image and standard picture to carry out Pixel-level computing detection technique in most of existing printed matter detection systems, therefore, must eliminate the difference of this two width image on the locus before computing, namely this two width image is carried out hi-Fix, the accuracy of guarantee operation result.In sum, the framing of printed matter full frame detection system is the important prerequisite condition of print quality parameter detecting, and its positioning accuracy is directly connected to the accuracy that mass parameter detects, and its locating speed directly has influence on the efficient of detection system.
Existing framing algorithm is a lot, be widely used in various fields, these location algorithms have certain similitude in theory, but because the view data complexity that location algorithm is processed is various, and each field is not quite similar to the requirement of algorithm, cause not blanket framing algorithm, concerning different image and application, its localization method that adopts may have very large difference.In addition, the framing algorithm of most to as if gray level image, therefore find the method that can efficiently process the coloured image orientation problem is the hot issue of studying in image processing field always.
For now, for just start to walk the at home research of character of printed matter full frame detection technique, there is gap aspect the TT﹠C system of print quality, still take the complete machine introduction as main.External research has entered the stage of comparative maturity, and oneself the printing equipment develop of combination oneself of many well-known offset press factory commercial cities has gone out printing images color detection and control system separately.In these systems, mostly adopt template matching method, sequential similarity detection matching method or size coupling etc. that the coloured image of continuous acquisition is positioned and separates, measure accuracy rate higher, velocity ratio is very fast.But on the whole, both at home and abroad in the research and product development aspect printed matter full frame detection technique, can realize the image of linear array photoelectrical coupler camera continuous acquisition printed matter is realized location and the technology of separating, cost is higher, equipment price is expensive, function is too simple, operates comparatively loaded down with trivial detailsly, and the scope of application is narrower.
Summary of the invention
The invention provides a kind of printed matter image position method, be intended to solve both at home and abroad in the research and product development aspect printed matter full frame detection technique, can realize the image of linear array photoelectrical coupler camera continuous acquisition printed matter is realized location and the technology of separating, cost is higher, equipment price is expensive, function is too simple, operate comparatively loaded down with trivial details, the problem that the scope of application is narrower.
The object of the present invention is to provide a kind of printed matter image position method, described detection method comprises:
Step 1: obtain photoelectrical coupler and gather the abundantest single channel image of information content in image;
Step 2: the pixel value of all pixel values of this channel image and this passage stock paper is poor separately, and gets the absolute value of difference;
Step 3: the statistics that adds up of the image by row, column after to absolute difference;
Step 4: successively the statistical value of adjacent two row, two row judged four borders up and down of image according to edge determination order and decision condition, and then the location of completing printing images.
Wherein, the packages printing image increases bar shaped location aid mark before printing after, when the location directly from step 3.
The invention provides a kind of printed matter image position method based on the cumulative statistic law of image absolute difference, compared with prior art, it detects in the research of online framing technology at the printed matter full frame, print speed printing speed and the photoelectrical coupler collected by camera speed of printing machine are considered as emphasis, realized when print speed printing speed and photoelectrical coupler collected by camera speed are not mated, also can accurately locate the printed matter image of continuous acquisition in roll web, sheet-fed printing and packages printing, practical.In addition, also overcome the situations such as printing images diversity and packages printing to the impact of framing based on the printed matter image position method of the cumulative statistic law of image absolute difference, positional accuracy is high, and locating speed is fast.
Description of drawings
The realization flow figure of the printed matter framing that Fig. 1 provides for the embodiment of the present invention;
Fig. 2 gathers image for the common printed product linear array of the roll web photoelectrical coupler that need to position;
The first width image of Fig. 3 for using Fig. 1 printed matter framing implementing procedure network for location 2 to obtain;
The second width image of Fig. 4 for using Fig. 1 printed matter framing implementing procedure network for location 2 to obtain;
Fig. 5 gathers image for the common printed product linear array of the sheet-fed photoelectrical coupler that need to position;
The first width image of Fig. 6 for using Fig. 1 printed matter framing implementing procedure network for location 5 to obtain;
The second width image of Fig. 7 for using Fig. 1 printed matter framing implementing procedure network for location 5 to obtain;
Fig. 8 is for using the printed matter image after aid mark is located in bar shaped;
Fig. 9 gathers image for the roll web package printed matter linear array photoelectrical coupler that need to position;
The first width image of Figure 10 for using Fig. 1 printed matter framing implementing procedure network for location 9 to obtain;
The second width image of Figure 11 for using Fig. 1 printed matter framing implementing procedure network for location 9 to obtain;
Figure 12 gathers image for the sheet-fed package printed matter linear array photoelectrical coupler that need to position;
The first width image of Figure 13 for using Fig. 1 printed matter framing implementing procedure location Figure 12 to obtain;
The second width image of Figure 14 for using Fig. 1 printed matter framing implementing procedure location Figure 12 to obtain;
The specific embodiment
In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, is not intended to limit the present invention.
Fig. 1 shows the realization flow figure of the printed matter image position method that the embodiment of the present invention provides.
The embodiment of the present invention provides a kind of printed matter image position method, provides two kinds of different implementation steps for common printed product and two kinds of situations of package printed matter.
As follows for the concrete implementation step in the location of common printed product image:
In step S101, use photoelectrical coupler to obtain the printed matter digital picture of continuous acquisition, and determine to gather the abundantest single channel image of information content in image.
In this step, determine the condition of the single channel image that information content is the abundantest: all pixel value sums in subchannel statistics passage, that passage of statistical value maximum is the abundantest single channel of information content; If maximum statistical value has when two or more, judge with the max pixel value of each passage and the extent of minimum pixel value, that passage of difference maximum is the abundantest single channel of information content.
In step S102, each pixel value in the single channel that information content is the abundantest is all done poorly with the pixel value of this single channel stock paper, and records the absolute value of difference according to the original coordinate position of consecutive image pixel.
In this step, the pixel value of stock paper be by photoelectrical coupler with the equivalent environment that gathers the printed matter consecutive image under, gather that the stock paper obtains.
In step S103, by row, by row to the absolute value that records in the step S102 statistics that add up, and record in order statistical value, and be worth corresponding line number or be listed as number.
In step S104, successively the statistical value of adjacent two row, two row judged four borders up and down of image according to edge determination order and decision condition, and then the location of completing printing images.
In this step, edge determination order and decision condition are as follows:
Web printing edge determination order and decision condition:
The coboundary decision condition:
Wherein R (i) is that wherein i represents every row line number of absolute difference image by the pixel accumulated value of row statistics absolute value images, and shor is the horizontal resolution of independent image before seal.When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain single image, represents with upi;
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the intrinsic vertical resolution of single image before seal
The left margin decision condition:
Wherein C (j) is the pixel accumulated value by the absolute value images behind row statistics up-and-down boundary location, and wherein j represents that every row row number, dver are the vertical resolutions through the image behind the up-and-down boundary location.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the row number of the right margin of certain independent image.
Sheet-fed printing edge determination order and decision condition:
The left margin decision condition:
Wherein C (j) is the pixel accumulated value by row statistics absolute value images, and wherein j represents every row row number of absolute value images, and ver is the vertical resolution of image to be positioned.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the row number of the right margin of certain independent image.
The coboundary decision condition:
R (i) is that wherein i represents every row line number of absolute difference image by the pixel accumulated value of the absolute difference image after the left and right boundary alignment of row statistics, and dhor is the horizontal resolution of the single independent image after the boundary alignment of left and right.When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain independent image, represents with upi.
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the vertical resolution of single independent image before seal.
As follows for the concrete implementation step in the location of package printed matter image:
In step S101 ', before the printed matter printing, locate aid mark in the surrounding increase bar shaped of its original copy.Then use photoelectrical coupler to obtain the printed matter digital picture of continuous acquisition, and determine to gather the abundantest single channel image of information content in image, and
In this step, the width that aid mark is located in bar shaped is 3 millimeters, need to be placed on the surrounding of being close to the printed matter original copy that will orient.Bar shaped location aid mark has 4 kinds of colors and can select, be respectively yellow, cyan, magenta and black, the rule of its choice for use is: the location aid mark of selecting to be with the printing images dominant hue monochrome ink color of complementary color relation, be that mass-tone is used the cyan mark when being adjusted to redness, when being adjusted to green, mass-tone uses the magenta mark, when being adjusted to blueness, mass-tone uses yellow mark, determining of dominant hue is mainly to judge by the RGB triple channel, and namely in passage, the cumulative and maximum passage of all pixel values is dominant hue; When the more difficult resolution of dominant hue, can consider to use comparatively general density bullet, if consider printing cost, use density bullet to compare and other monochromatic marks, the cost of its printing ink is more cheap.
In this step, determining of the single channel image that information content is the abundantest is according to the color of the bar shaped location aid mark that uses and fixed.If what use is yellow telltale mark, the single channel image that information content is the abundantest is the blue channel image; If what use is the cyan telltale mark, the single channel image that information content is the abundantest is red channel image; If what use is the magenta telltale mark, the single channel image that information content is the abundantest is green channel image; If what use is the black telltale mark, the red, green, blue channel image all can be elected to be the abundantest single channel image of information content.
In step S102 ', according to original this single pass each pixel value of coordinate position record of consecutive image pixel.
In step S103, by row, by row to the pixel value of record in the step S102 ' statistics that add up, and record in order statistical value, and be worth corresponding line number or be listed as number.
In step S104, successively the statistical value of adjacent two row or two row judged four borders up and down of image by edge determination order and decision condition, and then the location of completing the printed matter image.
In this step, edge determination order and decision condition are as follows:
If what use in the packages printing process is roll web, its edge determination order and decision condition are:
The coboundary decision condition:
Wherein R (i) is the cumulative sum of the capable image pixel value of i; I is the line number of the every row of image; Hor is the horizontal resolution of image to be positioned; When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain independent image, represents with upi.
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the vertical resolution of single independent image before seal;
The left margin decision condition:
Wherein C (j) is the cumulative sum of image pixel value of j row; J is the row number of the every row of image; Dver is the vertical resolution of the image after up-and-down boundary is located.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the row number of the right margin of certain independent image.
If what use in the packages printing process is sheet-fed, its edge determination order and decision condition are:
The left margin decision condition:
Wherein C (j) is the cumulative sum of image pixel value of j row; J is the row number of the every row of image; Ver is the vertical resolution of image to be positioned.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the row number of the right margin of certain independent image.
The coboundary decision condition:
Wherein R (i) is the cumulative sum of the capable image pixel value of i; I is the line number of the every row of image; Wherein dhor is the horizontal resolution of the single independent image behind the location, left and right.When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain independent image, represents with upi.
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the vertical resolution of single independent image before seal;
Below in conjunction with accompanying drawing and real case, technical scheme of the present invention is further described, it is more remarkable that purpose of the present invention and effect will become.
The embodiment of the present invention 1 provides a kind of printed matter image position method based on the cumulative statistic law of image absolute difference, and its objective for implementation is that the common printed product linear array of roll web photoelectrical coupler gathers image (Fig. 2).
At first the method calculates to determine to the printed matter image (Fig. 2) that photoelectrical coupler collects the single channel image that information content is the abundantest.By three passages of red, green, blue, Fig. 2 is added up, all pixel values in three passages are added respectively and, due to adding and value maximum of red passage, red channel image namely is confirmed as the abundantest single channel image of information content.
The second, extract the pixel value of paper in red channel image, the pixel value that red channel image is all and the pixel value of this paper carry out respectively the difference computing, and record the absolute value of these differences by the original coordinate position of consecutive image pixel.
The 3rd, sum up by being about to the absolute value of every row, and record each and add and be worth corresponding line number.Product common printed according to roll web coboundary Rule of judgment can be judged from top to bottom and count, the coboundary of the 157th independent printed matter image of behavior.Because the original vertical resolution ratio of this independent printed matter is 680, its lower boundary should gather at printed matter the capable place of 157+680 of image so, i.e. 837 row places.
The 4th, by row, the absolute value of every row is summed up, and record each and add and be worth corresponding row number.Product left margin Rule of judgment common printed according to roll web can be judged from left to right and count, and the 145th classifies the left margin of first independent printed matter image as.Product right margin Rule of judgment common printed according to roll web can be judged from left to right and count, and the 945th classifies the right margin of first independent printed matter image as.According to four borders up and down, can locate the first independent printed matter image of width, and cut, obtain this image (Fig. 3).By that analogy, can judge from left to right and count, the 1020th classifies the left margin of second independent printed matter image as, and the 2040th classifies the right margin of second independent printed matter image as.According to four borders up and down, can locate the second independent printed matter image of width, and cut, obtain this image (Fig. 4).By that analogy, can obtain independent printed matter image afterwards.
The embodiment of the present invention 2 provides a kind of printed matter image position method based on the cumulative statistic law of image absolute difference, and its objective for implementation is that the common printed product linear array of sheet-fed photoelectrical coupler gathers image (Fig. 5).
At first the method calculates to determine to the printed matter image (Fig. 5) that photoelectrical coupler collects the single channel image that information content is the abundantest.By three passages of red, green, blue, Fig. 5 is added up, all pixel values in three passages are added respectively and, due to adding and value maximum of red passage, red channel image namely is confirmed as the abundantest single channel image of information content.
The second, extract the pixel value of paper in red channel image, the pixel value that red channel image is all and the pixel value of this paper carry out respectively the difference computing, and record the absolute value of these differences by the original coordinate position of consecutive image pixel.
The 3rd, by row, the absolute value of every row is summed up, and record each and add and be worth corresponding row number.Product left margin Rule of judgment common printed according to sheet-fed can be judged from left to right and count, and the 145th classifies the left margin of independent printed matter image as.Product right margin Rule of judgment common printed according to sheet-fed can be judged from left to right and count, and the 945th classifies the right margin of first independent printed matter image as, and then the first independent printed matter image of width is cut out from consecutive image.By that analogy, can judge from left to right and count, the 1020th classifies the left margin of second independent printed matter image as, and the 2040th classifies the right margin of second independent printed matter image as, and then the second independent printed matter image of width is cut out from consecutive image.By that analogy, each the independent printed matter image with the back cuts out.
The 4th, sum up by the absolute value of every row of each independent printing images of being about to cut out, and record each and add and be worth corresponding line number.Product common printed according to sheet-fed coboundary Rule of judgment, can judge from the top down and count, the coboundary of the first independent printed matter image of width after the 157th behavior left and right boundary alignment, because the original vertical resolution ratio of this independent printed matter is 680, its lower boundary should gather at printed matter the capable place of 157+680 of image so, i.e. 837 row places, the first independent printed matter image of width after the boundary alignment of left and right is carried out cutting of up-and-down boundary, can obtain the first final independent printed matter image of width (Fig. 6); By that analogy, can judge from the top down and to count, the coboundary of the second independent printed matter image of width after the 125th behavior left and right boundary alignment is because the original vertical resolution ratio of this independent printed matter is 680, its lower boundary should gather at printed matter the capable place of 125+680 of image so, i.e. 805 row places.The second independent printed matter image of width after the boundary alignment of left and right is carried out cutting of up-and-down boundary, can obtain the second final independent printed matter image of width (Fig. 7); By that analogy, and then the independent printed matter image of every width after obtaining.
The embodiment of the present invention 3 provides a kind of printed matter image position method based on the cumulative statistic law of image absolute difference, and its objective for implementation is that roll web package printed matter linear array photoelectrical coupler gathers image (Fig. 9).
At first the method adds bar shaped location aid mark in the surrounding of package printed matter original copy, obtains Fig. 8.By using photoelectrical coupler to gather the roll paper printing product image of Fig. 8, obtain Fig. 9.What use due to Fig. 8 is cyan bar shaped location aid mark, therefore just must determine that red channel image is the abundantest single channel image of information content.
The second, record each pixel value of red channel image by the original coordinate position of consecutive image pixel.
The 3rd, sum up by being about to the pixel value of every row, and record each and add and be worth corresponding line number.According to roll web package printed matter coboundary Rule of judgment, can judge from top to bottom and count, the coboundary of the 196th independent printed matter image of behavior.Because the original vertical resolution ratio of this independent printed matter is 900, its lower boundary should gather at printed matter the capable place of 196+900 of image so, i.e. 1096 row places.
The 4th, by row, the pixel value of every row is summed up, and record each and add and be worth corresponding row number.According to roll web package printed matter left margin Rule of judgment, can judge from left to right and count, the 76th classifies the left margin of first independent printed matter image as.According to roll web right margin Rule of judgment, can judge from left to right and count, the 676th classifies the right margin of first independent printed matter image as.According to four borders up and down, can locate the first independent printed matter image of width, and cut, obtain this image (Figure 10).By that analogy, can judge from left to right and count, the 702nd classifies the left margin of second independent printed matter image as.According to roll web right margin Rule of judgment, the 1602nd classifies the right margin of second independent printed matter image as.According to four borders up and down, can locate the second independent printed matter image of width, and cut, obtain this image (Figure 11).By that analogy, can obtain independent printed matter image afterwards.
The embodiment of the present invention 4 provides a kind of printed matter image position method based on the cumulative statistic law of image absolute difference, and its objective for implementation is that sheet-fed package printed matter linear array photoelectrical coupler gathers image (Figure 12).
At first the method adds bar shaped location aid mark in the surrounding of package printed matter original copy, obtains Fig. 8.By using photoelectrical coupler to gather the sheet-fed printed matter image of Fig. 8, obtain Figure 12.What use due to Fig. 8 is cyan bar shaped location aid mark, therefore just must determine that red channel image is the abundantest single channel image of information content.
The second, record each pixel value of red channel image by the original coordinate position of consecutive image pixel.
The 3rd, by row, the pixel value of every row is summed up, and record each and add and be worth corresponding row number.According to sheet-fed package printed matter left margin Rule of judgment, can judge from left to right and count, the 76th classifies the left margin of independent printed matter image as.According to sheet-fed package printed matter right margin Rule of judgment, can judge from left to right and count, the 676th classifies the right margin of first independent printed matter image as, and then the first independent printed matter image of width is cut out from consecutive image.By that analogy, can judge from left to right and count, the 702nd classifies the left margin of second independent printed matter image as, and the 1602nd classifies the right margin of second independent printed matter image as, and then the second independent printed matter image of width is cut out from consecutive image.By that analogy, each the independent printed matter image with the back cuts out.
The 4th, sum up by the pixel value of every row of each independent printing images of being about to cut out, and record each and add and be worth corresponding line number.According to sheet-fed package printed matter coboundary Rule of judgment, can judge from the top down and count, the coboundary of the first independent printed matter image of width after the 196th behavior left and right boundary alignment, because the original vertical resolution ratio of this independent printed matter is 900, its lower boundary should gather at printed matter the capable place of 196+900 of image so, i.e. 1096 row places, the first independent printed matter image of width after the boundary alignment of left and right is carried out cutting of up-and-down boundary, can obtain the first final independent printed matter image of width (Figure 13); By that analogy, can judge from the top down and to count, the coboundary of the second independent printed matter image of width after the 98th behavior left and right boundary alignment is because the original vertical resolution ratio of this independent printed matter is 900, its lower boundary should gather at printed matter the capable place of 98+900 of image so, i.e. 998 row places.The second independent printed matter image of width after the boundary alignment of left and right is carried out cutting of up-and-down boundary, can obtain the second final independent printed matter image of width (Figure 14); By that analogy, and then the independent printed matter image of every width after obtaining.
The printed matter image position method that the embodiment of the present invention provides, compared with prior art, it detects in the research of online framing technology at the printed matter full frame, print speed printing speed and the photoelectrical coupler collected by camera speed of printing machine are considered as emphasis, realized when print speed printing speed and photoelectrical coupler collected by camera speed are not mated, also can accurately locate the printed matter image of continuous acquisition in roll web, sheet-fed printing and packages printing, practical.In addition, also overcome the situations such as printing images diversity and packages printing to the impact of framing based on the printed matter image position method of the cumulative statistic law of image absolute difference, positional accuracy is high, and locating speed is fast.
The above is only preferred embodiment of the present invention, not in order to limiting the present invention, all any modifications of doing within the spirit and principles in the present invention, is equal to and replaces and improvement etc., within all should being included in protection scope of the present invention.
Claims (9)
1. the localization method to continuous acquisition printed matter image, is characterized in that, described method comprises:
Step 1: the use photoelectrical coupler obtains the printed matter of continuous acquisition, and the abundantest single channel image of definite information content;
Step 2: the abundantest single channel image of first determination information amount, and each pixel value of the single channel image that information content is the abundantest is all poor separately with the pixel value of this passage stock paper, and records the absolute value of each difference by the original coordinate position of consecutive image pixel;
Step 3: by row, column in step 2 in the absolute value at the original coordinate position of consecutive image pixel place or the pixel value statistics that adds up, and record in order line number corresponding to statistical value or row number;
Step 4: successively the statistical value of adjacent two row, two row judged four borders up and down of image according to edge determination order and decision condition, and then the location of completing printing images.
2. method according to claim 1, it is characterized in that, when described printed matter is package printed matter, need after the surrounding of printed matter original copy adds that aid mark is located in bar shaped, re-use the printed matter that photoelectrical coupler obtains continuous acquisition, and the abundantest single channel image of definite information content.
3. method according to claim 1, it is characterized in that, when described printed matter is package printed matter, step 2 item becomes directly each pixel value by the abundantest single channel image of the original coordinate position recorded information amount of consecutive image pixel, and each pixel value of single channel image that need not information content is the abundantest is all poor separately with the pixel value of this passage stock paper.
4. method according to claim 1, is characterized in that, the decision method of the single channel image that described information content is the abundantest is: all pixel value sums in subchannel statistics passage, and that passage of statistical value maximum is the abundantest single channel of information content; If maximum statistical value has when two or more, judge with the max pixel value of each passage and the extent of minimum pixel value, that passage of difference maximum is the abundantest single channel of information content.
5. method according to claim 2, it is characterized in that, the width of described bar shaped location aid mark is 3mm, its color has yellow, magenta, four kinds of cyan and black, can select the bar shaped of different colours to locate aid mark according to the feature of printing images and be placed on the surrounding of printing images with accurate location, its color selective rules is as follows: the location aid mark of selecting to be with the printing images dominant hue monochrome ink color of complementary color relation, be that mass-tone is used the cyan mark when being adjusted to redness, when being adjusted to green, mass-tone uses the magenta mark, when being adjusted to blueness, mass-tone uses yellow mark, determining of dominant hue is mainly to judge by the RGB passage, be that in passage, the cumulative and maximum passage of all pixel values is dominant hue, when the more difficult resolution of dominant hue, can consider to use comparatively general density bullet, if consider printing cost, use density bullet to compare and other monochromatic marks, the cost of its printing ink is more cheap.
6. method according to claim 1, is characterized in that, described edge determination order and decision condition are according to the difference of web printing, sheet-fed printing and these three kinds of applicable cases of packages printing and difference.
7. method according to claim 6, is characterized in that, judgement order and the decision condition on four borders up and down of described web printing image are as follows:
The coboundary decision condition:
Wherein R (i) is that wherein i represents every row line number of absolute difference image by the pixel accumulated value of row statistics absolute value images, and shor is the horizontal resolution of independent image before seal.When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain single independent image, represents with upi;
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the intrinsic vertical resolution of single image before seal.
The left margin decision condition:
Wherein C (j) is the pixel accumulated value by the absolute value images behind row statistics up-and-down boundary location, and wherein j represents that every row row number, dver are the vertical resolutions through the image behind the up-and-down boundary location.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the row number of the right margin of certain independent image.
8. method according to claim 6, is characterized in that, judgement order and the decision condition on four borders up and down of described sheet-fed printing images are as follows:
The left margin decision condition:
Wherein C (j) is the pixel accumulated value by row statistics absolute value images, and wherein j represents every row row number of absolute value images, and ver is the vertical resolution of image to be positioned.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the right margin of certain independent image.
The coboundary decision condition:
R (i) is that wherein i represents every row line number of absolute difference image by the pixel accumulated value of the absolute difference image after the left and right boundary alignment of row statistics, and dhor is the horizontal resolution of the single independent image after the boundary alignment of left and right.When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain independent image, represents with upi.
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the vertical resolution of single independent image before seal.
9. method according to claim 6, is characterized in that, judgement order and the decision condition on four borders up and down of described packages printing image are as follows:
If what use in the packages printing process is roll web, the judgement on four borders up and down of its image order and decision condition are as follows:
The coboundary decision condition:
Wherein R (i) is the cumulative sum of the capable image pixel value of i; I is the line number of the every row of image; Hor is the horizontal resolution of image to be positioned; When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain independent image, represents with upi.
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the vertical resolution of single independent image before seal;
The left margin decision condition:
Wherein C (j) is the cumulative sum of image pixel value of j row; J is the row number of the every row of image; Dver is the vertical resolution of the image after up-and-down boundary is located.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the row number of the right margin of certain independent image.
If what use in the packages printing process is sheet-fed, the judgement on four borders up and down of its image order and decision condition are as follows:
The left margin decision condition:
Wherein C (j) is the cumulative sum of image pixel value of j row; J is the row number of the every row of image; Ver is the vertical resolution of image to be positioned.When certain j satisfied this condition, this j was exactly the row number of the left margin of certain independent image.
The right margin decision condition:
When certain j satisfied this condition, this j was exactly the row number of the right margin of certain independent image.
The coboundary decision condition:
Wherein R (i) is the cumulative sum of the capable image pixel value of i; I is the line number of the every row of image; Wherein dhef is the horizontal resolution of the single independent image behind the location, left and right.When certain i satisfied this condition, this i was exactly the line number of the coboundary of certain independent image, represents with upi.
Thus, the line number of the lower boundary of certain single independent image is upi+sver,
Wherein sver is the vertical resolution of single independent image before seal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100117063A CN103085471A (en) | 2013-01-11 | 2013-01-11 | Positioning method for continuous acquisition of presswork images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100117063A CN103085471A (en) | 2013-01-11 | 2013-01-11 | Positioning method for continuous acquisition of presswork images |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103085471A true CN103085471A (en) | 2013-05-08 |
Family
ID=48198821
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100117063A Pending CN103085471A (en) | 2013-01-11 | 2013-01-11 | Positioning method for continuous acquisition of presswork images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103085471A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115753812A (en) * | 2022-11-01 | 2023-03-07 | 东方合智数据科技(广东)有限责任公司 | Printing quality detection method and related equipment in paper packaging industry |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020122175A1 (en) * | 2001-01-10 | 2002-09-05 | Dance Christopher R. | Document optimized reconstruction of color filter array images |
| CN101244649A (en) * | 2008-03-28 | 2008-08-20 | 北京工业大学 | Automatic detection method for printed product four-color register partial difference |
| CN102221554A (en) * | 2011-04-06 | 2011-10-19 | 天津科技大学 | Presswork quality and flaw inspection method |
| CN102253042A (en) * | 2011-04-06 | 2011-11-23 | 天津科技大学 | Dot area rate detection method of printed product |
-
2013
- 2013-01-11 CN CN2013100117063A patent/CN103085471A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020122175A1 (en) * | 2001-01-10 | 2002-09-05 | Dance Christopher R. | Document optimized reconstruction of color filter array images |
| CN101244649A (en) * | 2008-03-28 | 2008-08-20 | 北京工业大学 | Automatic detection method for printed product four-color register partial difference |
| CN102221554A (en) * | 2011-04-06 | 2011-10-19 | 天津科技大学 | Presswork quality and flaw inspection method |
| CN102253042A (en) * | 2011-04-06 | 2011-11-23 | 天津科技大学 | Dot area rate detection method of printed product |
Non-Patent Citations (3)
| Title |
|---|
| HAO JIANQIANG ET AL: "Research of Image Location for Print Whole Image On-line Detection", 《APPLIED MECHANICS AND MATERIALS》 * |
| 李铭祺等: "印刷图像识别和定位技术的研究", 《微计算机信息》 * |
| 郝健强等: "基于相位相关的印刷全画面检测中图像定位的研究", 《包装工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115753812A (en) * | 2022-11-01 | 2023-03-07 | 东方合智数据科技(广东)有限责任公司 | Printing quality detection method and related equipment in paper packaging industry |
| CN115753812B (en) * | 2022-11-01 | 2023-08-18 | 东方合智数据科技(广东)有限责任公司 | Printing quality detection method and related equipment in paper packaging industry |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102253042B (en) | Dot area rate detection method of printed product | |
| CN101244649B (en) | Automatic detection method for printed product four-color register difference | |
| CN102811863A (en) | Automated Inspection Of A Printed Image | |
| CN102221554A (en) | Presswork quality and flaw inspection method | |
| CN110378179B (en) | Subway ticket evasion behavior detection method and system based on infrared thermal imaging | |
| CN102221338B (en) | Method for detecting thickness of ink layer of presswork | |
| CN111242896A (en) | Color printing label defect detection and quality rating method | |
| CN102842203B (en) | Method for monitoring bridge fault on basis of video image | |
| CN102765249A (en) | Machine vision detection method based on four-colour printed matter registration detection marks | |
| CN103954634A (en) | Online quality detection system for printed matter | |
| CN108162600A (en) | The method and test pattern of failure print nozzles in detection and compensation ink-jet printer | |
| CN100583147C (en) | Method for evaluating the quality of a printed matter, produced by a printing machine | |
| CN102183224B (en) | Method for detecting screen dot of measurement and control strip based on machine vision | |
| CN108009556B (en) | River floating object detection method based on fixed-point image analysis | |
| CN109911550A (en) | Scratch board conveyor protective device based on infrared thermal imaging and visible light video analysis | |
| CN105083912A (en) | Conveyer belt deviation detecting method based on image identification | |
| CN113873229B (en) | Image dead pixel detection method, system and device | |
| CN105136811B (en) | OVMI optics gradual change magnetic ink printing product on-line detecting system and method | |
| CN103837537B (en) | A kind of without Grains metal inkjet forme dot area percentage measuring method | |
| CN105083913A (en) | Conveyer belt tearing detecting method based on machine visions | |
| CN102236925A (en) | System and method for offline secondary detection and checking of machine detected data of large-piece checker | |
| CN111766248A (en) | Steel seal on-line detection system and method based on color CCD | |
| CN102749034A (en) | Railway switch gap offset detection method based on image processing | |
| CN103085471A (en) | Positioning method for continuous acquisition of presswork images | |
| CN101425143B (en) | Image positioning method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C05 | Deemed withdrawal (patent law before 1993) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130508 |