Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/70—Determining position or orientation of objects or cameras
  • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
  • G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
  • G06V10/00—Arrangements for image or video recognition or understanding
  • G06V10/20—Image preprocessing
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
  • G06V10/00—Arrangements for image or video recognition or understanding
  • G06V10/20—Image preprocessing
  • G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• • G—PHYSICS
  • G07—CHECKING-DEVICES
  • G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
  • G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
  • G07D7/20—Testing patterns thereon
  • G07D7/202—Testing patterns thereon using pattern matching
  • G07D7/206—Matching template patterns

Definitions

• the invention relates to methods for the qualitative assessment of material according to the preamble of claim 1, 10 or 11.
• a method is known for example from DE 196 13 082 C2.
• the printed material for example sheets printed with banknotes or banknotes, which are provided with a window thread, hologram or kinegram, is illuminated with an illumination device such that the light reflected by the printed material falls into a photoelectric sensor.
• the image thus recorded by the photoelectric sensor can then be evaluated in an evaluation device, for example a standard computer with suitable evaluation software, and checked for printing errors.
• a disadvantage of the known methods is that recognition features whose image properties do not deviate sufficiently from the image properties of the other printed image after being recorded by the photoelectric sensor, as is the case, for example, with colored window threads, are not recognized with sufficient reliability by the evaluation device can.
• DE 10000 364 A1 describes a method for the feature-based determination of errors. An average background level is calculated and separated to reinforce properties of certain features.
• DE 26 20767 C2 discloses a method for checking the print quality of printed images, the partial images originating from different printing methods being combined to form an overall template.
• US 5 125 037 A discloses a method for checking printed images, wherein additional marks are printed. The position and density of these print marks is checked.
• the invention has for its object to provide methods for the qualitative assessment of material.
• the object is achieved by the features of claims 1, 10 or 11.
• the invention is based on the basic idea that, when evaluating position-variant identification features in which the optical properties, for example the reflectivity, are not sufficient for sufficiently reliable identification, known information about these identification features is also incorporated into the evaluation. It is assumed as a premise that the position-variant identification features, for example a colored window thread, differ at least in partial areas in the optical properties, for example in the gray value, so far from the other material to be inspected, e.g. B. Distinguish the printed image that at least there is no complete match between the identification feature and the printed image.
• One advantage of the method is, in particular, that additional information about the known geometric contour of the identification feature or the relative arrangement of several identification features present in the printed image is evaluated for determining the position of the identification variant with position variation.
• This additional data is stored in a mask reference, which is stored for each material to be evaluated, and which represents the geometric data in a suitable form.
• a background reference value is stored in the evaluation device, which represents the optical properties of the print image in at least a part of a surrounding area that surrounds the identification feature.
• the properties of the background reference value must differ at least slightly from the optical properties of the identification feature to be identified.
• the current image data and the background reference value then become a difference image, at least for the Area of expectation formed.
• the difference image essentially all features of the print image are hidden by difference formation, which correspond in their optical properties to the background reference value. Only the areas of the position-variant identification feature and also other elements, such as printing errors or edge deviations, are depicted in the difference image due to their deviation from the background reference value, the areas of the position-variant identification feature having particularly high amplitudes.
• the difference image data is compared with the mask reference and the current position of the recognition feature is inferred from the result of the comparison.
• This method step is based on the consideration that the difference image is essentially determined by the inclusion of the position-variant identification feature, so that the actual position of the position-variant identification characteristic can be inferred from extensive coverage between the mask reference and the difference image. If it is not possible to determine sufficient coverage between the mask reference and the differential image data due to other error influences, this is harmless, since this only leads to an error display in the print image control and to the removal of the corresponding sheet.
• the areas of the printed image that result from the current position of the identification feature are hidden during the subsequent qualitative assessment of the printed material, so that disruptions in the inspection of the printed image due to the position variant arrangement of the identification feature are excluded.
• the recognition of the position-variant recognition feature when the method is carried out can be improved further in that a binarization threshold is stored in the evaluation device. After from the current image data and the If the background reference value the difference image was formed, then all image data whose values are below the binarization threshold can then be filtered out from the difference image. That is, Only those image points are preserved in the difference image that differ sufficiently significantly from the normal printed image so that the usually other deviations, for example printing errors or edge deviations, can be hidden from the difference image.
• the procedure can be such that the mask reference is shifted until a maximum overlap between the mask reference and the difference image results.
• Various mathematical evaluation methods can be used to evaluate the coverage between the mask reference and the difference image in order to find the corresponding coverage maximum.
• the overlap between the difference image and the mask reference should be calculated in data form if possible based on suitable mathematical operations.
• One way of evaluating the coverage between the mask reference and the difference image is to calculate centers of gravity according to the optical distribution of the pixels in the difference image and to compare these centers of gravity with the center of gravity of the mask reference. A maximum coverage occurs when the sum of the center of gravity differences between the mask reference and the difference image is minimized.
• a prerequisite for the implementation of the method is the storage of a suitable background reference value in the evaluation device.
• the background reference value for example based on empirical values, can simply be specified as a process parameter. However, it is advantageous if the background reference value is specifically determined in a learning phase depending on the respective print image of the material to be inspected. Two alternatives are given below.
• printed reference material is used in the learning phase, which does not contain the position variant identification feature.
• sheets printed with value notes can be used for which the window thread is not present.
• the background reference value can be derived by a suitable evaluation of this reference material without a distinguishing feature.
• the learning phase can also be carried out with reference material that contains the position-variant identification feature. If, during the evaluation of the print image of the reference material, the position-variant identification features appear bright in comparison to the surrounding area, a threshold value is selected as the background reference value which corresponds to the values of the darkest pixels of the identification feature. During the subsequent assessment of the printed material, it is then assumed, based on the threshold value, that at least in the expected range, all pixels that are darker than the background reference value do not belong to the position-variant identification feature. If, on the other hand, the identification feature appears dark in comparison to the surrounding area, a threshold value is selected as the background reference value, the value of which corresponds to the brightest pixels of the identification feature.
• Figure 1 is a schematically illustrated difference image in a view from above.
• FIG. 2 shows the difference image according to FIG. 1 after binarization has been carried out
• FIG. 3 shows the mask reference for determining the position of the position-variant identification feature in the difference image according to FIG. 2;
• FIG. 4 shows the overlap between the difference image according to FIG. 2 and the mask reference according to FIG. 3;
• FIG. 5 shows a second mask reference in a schematically represented side view
• Fig. 6 shows a second difference image in a schematically illustrated side view.
• the difference image 01 according to FIG. 1 was formed during the inspection of sheets printed with value notes, only a section of the difference image being shown in the area of a bank note in FIG. 1. It can be seen in FIG. 1 that the normal printed image of the banknote is hidden in the differential image 01 and only the areas of the printed image which differ significantly from the background reference value are depicted as dark fields in the differential image.
• the position of a window thread 03 incorporated in the printed sheet, which corresponds to itself in the difference image 01, can be in a striped expectation area 02 indicated by dashed lines Breakthroughs in five dark fields 03 depicted vary.
• FIG. 3 shows a mask reference 06 in its geometric form.
• the data for the width 07 and the length 08 of the window thread openings are stored in the mask reference 06.
• the mask reference 06 also stores the values for the distance 09 between the window thread openings and the number of window thread openings per bank note.
• the mask reference 06 is shifted relative to the difference image 01 by data technology operations until there is a maximum overlap between the mask reference 06 and the dark fields 03 in the difference image 01. If this maximum of coverage is reached, the distances 11; 12, e.g. B. the current positions in the X and Y directions of the mask reference 06 relative to the edges of the banknote on the current position of the window thread 03 in the print image, so that the areas of the window thread openings can be hidden in a subsequent inspection of the print image.
• FIG. 5 shows a second mask reference 13, the dark fields 14 corresponding to eight window thread openings 14 during the inspection of a bank note on a concave represents curved contact surface.
• Fig. 6 shows a difference image 16 schematically, in which the window thread openings 14 in dark fields 17, z. B. in window threads 17 have shown.
• the dark field 18 was caused by a printing error 18 and not by a window thread opening 14.
• a window thread opening 14 in the middle did not appear in the difference image 16 due to the insufficient color difference between the substrate and window thread 17.
• mask reference 13 is projected onto a projection line 19 and the resulting light-dark distribution with the light-dark distribution resulting from the projection of the difference image 16 onto a projection line 21 compared.
• This one-dimensional comparison of the light-dark distribution enables the position of the window thread 17 to be determined in one direction.

Applications Claiming Priority (3)

Publications (1)

Family

ID=7690703

Family Applications (1)

Country Status (7)

Families Citing this family (8)

Family Cites Families (20)

• 2001
  • 2001-07-05 DE DE10132589A patent/DE10132589B4/de not_active Expired - Lifetime
• 2002
  • 2002-06-25 US US10/482,159 patent/US7184584B2/en not_active Expired - Lifetime
  • 2002-06-25 AU AU2002354816A patent/AU2002354816A1/en not_active Abandoned
  • 2002-06-25 WO PCT/DE2002/002315 patent/WO2003005290A2/de active Application Filing
  • 2002-06-25 JP JP2003511185A patent/JP3760446B2/ja not_active Expired - Fee Related
  • 2002-06-25 CN CNB028133730A patent/CN1271563C/zh not_active Expired - Fee Related
  • 2002-06-25 EP EP02750805A patent/EP1417635A2/de not_active Ceased

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events