DE19632058C1 - Opto electronic identification or target element - Google Patents

Abstract

The element has a disc-shaped recognition mark (2) with a circular contour and a number of circular code elements (4), positioned on the outside of the recognition mark along concentric lines. The code elements are smaller in diameter than the recognition mark.They are positioned in accordance with a ring code relative to the centre of the recognition mark. The recognition mark is identified by a further marking, e.g. a cross or a point.

Description

The invention relates to an optoelectronically detectable Identification or target element according to the generic term of Claim 1 and a method for optoelectronic detection of such an identification or target element.

A protruding element can, for example, be a target element in photogrammetric three-dimensional measurement, so-called 3-D measurement, find use in the Machine tool measurement, when determining structural deformations, for example in buildings or Machine parts, is applied. It has been here ever since required a test object with a variety of  Image recording devices or or or from a variety of Take directions, with the image sections to Example must cover up to a third, with each one too measuring object point depicted in at least two images be and the assignment of the pixels to the object point must be created in order to carry out a photogrammetric measurement enable.

In the article "Industrial Photogrammetry: New Developments and Recent Applications "from Photogrammetric magazine Record 12 (68), pp. 197-217 dated October 1986 is on it pointed out that by using punctiform, Dog tags forming target elements are an object optoelectronically recorded and automatically digitally measured can be (digital photogrammetry). The identification of dog tags as such, the assignment of homologous Points in a variety of captured images with different However, camera configurations etc. are problematic viewed. However, it only uses uncoded Dog tags described as target marks.

An element mentioned above can also be used as Identification element for example for the General cargo identification and / or sorting, especially in Logistics (transport, warehousing, etc.) will.  

From DE 30 10 112 A1 is a piece goods identification element to be applied known with a circular structure that serves as a dog tag and from a scanning oscillating laser beam photoelectric reader is detectable. Inside the annular structure is alternating in shape, in Circumferential direction of different widths, black Circular disc segments provided a barcode by a a laser beam describing a circular path can be scanned. The code information stored therein is analyzed by analyzing the Gray value changes along the circular path of the Laser beam won. Identification of the dog tag however, only works if additional bad marks are sufficiently small. The photographic capture and Evaluation of the identification element is larger Inclination of the illumination axis to the surface of the element not possible anymore.

In the publication vol. 27, part B5, commission V des Committees for the 16th International Congress of International Society for Photogrammetry and Remote sensing (pages 647-656) becomes a technique described in the case of circular contours to detect a Identification or target element can be used. Here circular image structures are determined and with a Model circle compared; the most circular Image structure is identified as an identification tag. Of the  circular dog tag is a linear barcode assigned, which is then sampled in the linear Direction must be read. The image recording device must arranged vertically above the circular dog tags to ensure their circular mapping.

From publication vol. 29, part B5, commission V of Committees for the 17th international congress for Photogrammetry and remote sensing (pages 222-229) is a Known identification element whose identification tag from a circular disc with a concentric one Ring exists. The identification mark is determined in that an image structure with two circular objects with the same Center in template technology is sought, which in turn is an image capture in almost perpendicular direction to the circular plane required. Concentric to the dog tag center is a ring code around the Dog tag provided around. The coding is in the Quantity of the circular ring pieces in the circumferential direction. For extraction the code information is an analysis of the gray value change in Performed circumferential direction, with half of the circle to Referencing the code information is what the feasible code word length greatly reduced. There is none Possibility to check the accuracy of the Code information.

The task is an identification or target element to provide which regardless of the orientation of the Recording direction and from the recording distance optoelectronic can be detected and identified so that errors in photogrammetric measurement be excluded when identifying the target elements can or only minimal when sorting general cargo Requirements for the provision of the to be sorted Goods and the orientation of the Identification elements are placed on the goods have to.

This task is done through an identification or Target element of the type mentioned in the preamble of claim 1 solved, which is characterized according to the invention by the Dog tag around on at least one concentric Circular line encoded circular disc-shaped Coding elements.

So there exist for each identification or Target element predetermined positions by a Coding element can be occupied. The code information is that is, in the order in which these specific positions are allocated laid down.

The identification tag could be in the form of a circular ring be; in contrast, it has proven advantageous if the dog tag from a circular disc  Dot mark, which is characterized by preferably strong contrast stands out against the background.

The coding elements could be within a circular or circular disk-shaped identification tag are provided or they are in their particular positions around the Dog tag arranged around.

According to a particularly advantageous embodiment of the Invention, the coding elements are circular disks and preferably smaller than that Identification tag formed, as a result of which the identification or Target element more compact and the number of Coding options per area is increased. The Processing speed when recognizing an identification elements will be raised.

The coding elements can be on one or more Dog tag center arranged in concentric circles be. The possibility then opens up to the specific ones Positions on which coding elements are arranged can, for example in a star shape around the Arrange dog tag center around so that the Coding elements are provided at angular positions that differ only slightly, for example each at an angle of 5-15 degrees.  

The dog tag can have an additional one at its center Mark, such as a point or a cross, around to support other measurement types.

The present invention also has the object based on a method for optoelectronic detection and Processing the coding of an identification or Target element with a circular circumferential structure identification tag and with a ring code certain positions regarding the dog tag center to provide the intended coding elements, where a Object with identification or target element a digital image recording device is imaged.

This further task is accomplished through a procedure with the Features of claim 4 solved. Advantageous further training result from the dependent claims.

The The inventive method offers the advantage that an object can also be shot with the camera angle at an angle can. The circular perimeter structure of the dog tags then appears in the digital image as an elliptical image structure. In order to be able to recognize whether this is actually a Identification tag and the code information in the form of the To determine the arrangement of code elements, the Orientation of the elliptical image structure detected, so that on the underlying circular structure is closed can be. By one in the most general form as a rule or Condition to be designated query can be checked whether there is the elliptical structure is actually one Dog tag acts and whether that in the area of elliptical Structure other structures actually from a picture of the coding elements originate or whether it is too eliminating foreign objects. If it is certain that one ID tag is found, the coding is done by Determine the occupancy of the specific positions by Coding elements determined and the identification or Target element can be identified.

Further details and advantages of the invention result from the graphic representation and following description of an inventive Identification or target element and the inventive method. The drawing shows:

Figures 1a to 1e five different embodiments of an identification of the invention or target element.

Fig. 1f several identification or target elements according to Figure 1b with different coding.

Figure 2a is a description of an ellipse having five parameters.

Figure 2b is an illustration of a appearing in the digital image as an ellipse identification or target element.

2c, the reconstructed representation of the actual structure of the Identification- or target element according to Fig. 2b. and

Fig. 3 is a schematic representation of a flow chart.

Fig. 1 shows five different embodiments of an identification of the invention or target element. This consists of a circular disk-shaped identification mark, hereinafter referred to as a point mark 2 , and circular disk-shaped coding elements 4 provided at different positions around the point mark 2 , hereinafter referred to as coding points 4 . From the number of coding points arise in Figs. 1a to e code word lengths of 12 bits, 16 bits, 24 bits, 36 bits and 54 bits. In the embodiment according to FIG. 1a, the coding points are arranged on a circular line running concentrically around the identification mark center at angular positions which differ from one another by 30 °. In the star-shaped arrangement according to FIG. 1b, the angular positions differ by 360 ° / 16, and in FIGS. C and d by 360 ° / 24 and 10 °. The predetermined positions at which coding points can be provided are therefore evenly distributed in these embodiments. In the embodiment according to FIG. 1e, three independent arrangements of coding points on three concentric circular lines at angular positions are provided, which differ from one another by 30 °, 20 °, 360 ° / 24. The three arrangements can be distinguished by their respective radius of the circular line, and can be connected one after the other to identify the point mark and ultimately form a longer code word length. Further arrangements are conceivable.

Finally, FIG. 1f shows a plurality of identification or target elements of the embodiment according to FIG. 1b provided with identification numbers, each with different coding. The assignment order of the determined positions by the coding elements smaller than the point mark serves as decryption information for identifying the element.

The point mark is recorded by perspective imaging of the object by a digital image recording device and appears in the digital image as an elliptical structure ( FIG. 2). With image processing methods, all contours of the digital image are now preferably extracted and the contour coordinates, preferably all image structures recognized on the basis of their contrast, are calculated. From the orientation of the point mark appearing elliptical in the digital image and the arrangement of the coding points, the code and thus the identification number of the identification or target element can be determined independently of scale, rotation and observation position in further calculation steps.

Such a method is described below with reference to FIGS. 2a to c. Image structures are extracted using a contour determination algorithm, and the structure area and the structure circumference are calculated for these image structures. The different image coordinate scaling in the X and Y directions of the digital image recording device is taken into account. The shape factor for each image structure is preferably calculated from the circumference and the area according to the following equation.

For a circular structure, the form factor takes a minimum of 1; for other image structures, the value of the form factor is greater than 1. All image structures that have a form factor above a threshold value of, for example, 2 are classified as foreign objects or at least classified as an unlikely image structure for an image mark and therefore come at most as a possible image structure for coding points Question. Further differentiation criteria are minimum and maximum values for the scope of the image structures, which are determined with a priori knowledge of the geometric construction data of the identification or target element and the image recording conditions. In this way, elliptical image structures can be identified in the digital image as such or as a probable image structure for a point mark, the form factor of which is closer to the form factor 1 of a circle and the extent of which is in a certain size range. In this way, elliptical image structures are identified in the digital image and only the image structures that are likely to be a point mark are included in the further checks for finding and identifying the point mark. An image structure whose size is too large to be considered as a point mark or too small to be considered as a coding point is eliminated from the picture structure list since it cannot originate from a searched point mark or from coding points.

An ellipse can be described by five parameters, the center coordinates X₀, Y₀, the half-length of the major and minor axes a, b and the angle of rotation α, as shown in FIG. 2a. These parameters can be calculated from an ellipse adjustment with contour coordinates, or from the area moments of zero, first or second order. These five parameters are calculated for the image structures classified as point marks. For the structures of the digital image classified as coding points in the vicinity of the point mark, only the center point coordinates X _c , Y _{c are} determined by calculation of the center of gravity.

By determining the above-mentioned parameters for an elliptical image structure, their orientation is determined at the same time if one assumes that the image structure appearing elliptical in the digital image originates from a circular object structure. From this, the actual arrangement of further structures around the point mark can also be determined and, after a check, it can be determined whether these are the coding points of the identification or target element. For this purpose, a coordinate transformation is carried out, which can be understood using FIGS. 2a, b and c. A coordinate system x ′, y ′ is determined, which is rotated by the angle α relative to the image coordinate system X, Y of the image recording device and whose coordinate axes run along the main axes of the ellipse of the digital image ( FIGS. 2a, b). This coordinate system x ', y' is then rotated about a coordinate axis, so that a coordinate system x '', y '' is obtained in which the elliptical structure in the digital image has a circular shape ( Fig. 2c).

The distance r and the angular position θ of a coding point with respect to the point mark center can be calculated taking into account an image coordinate scaling factor s as follows (see FIGS. 2a, b, c).

The function atan2 (y _c ′ ′, x _c ′ ′) means the arctan function of (y _c ′ ′ / x _c ′ ′). If the image recording device has no image coordinate scaling difference or this should not be taken into account, then s = 1 is set. The coding points around the point mark for the identification or target element to be determined lie within a ring-shaped area in the actual local coordinate system x ′ ′, y ′ ′, and the following applies:

In addition, the scope of the coding point should be compared to the Scope of the point mark a minimum, maximum condition fulfill:

The criteria according to equations (6) and (7) are scale-invariant. The four constants k _r1 , k _r2 , k _u1 and k _u2 are to be determined from the geometric construction data of the point mark, taking into account the image measurement error. If the coding points are arranged in several rows around the point mark ( FIG. 1b, c, d) in order to increase the number of coding points, k _{r2 should be} correspondingly large.

With the equations from (2) to (5) and the criteria from (6) and (7) all coding points around the dot mark found and their angular position calculated around the point mark. To find the coding points around the point mark not just the objects classified as coding points,  but also the image structures classified as point marks included in the calculation as possible coding points.

After the best fit and discretization of the This results in angular positions at the defined positions binary ring code. The ring code corresponds to the binary code additional property that the resulting Identification number regardless of the rotational position of the Is binary information. Reduced with the same code word length However, the number of identification options for the ring code.

The total number of ring codes set should be Meet minimum, maximum and parity conditions. Furthermore should have at least one defined assignment sequence can be found within this ring code, so partially eliminate hidden dot marks. If all of the above Conditions are met, the point mark is considered valid accepted. For everyone classified as a point mark This validation is repeated for image structures.

The identification number for the identification or The target element results from the interpretation of the Ring codes. However, the interpretation can be various regulations can be implemented. The In any case, the interpretation result should be variable in rotation be independent of the observation position and of  the rotation of the element the identification number of the Dot mark remains consistent.

The number of coding combinations depends on the code word length of the ring code. The realizable code word length with the condition of equation (6) is limited by the resolution and accuracy of the angular position determination of the coding point around the point mark ( FIGS. 1a to d). With multiple conditions of equation (6) with different value pairs of k _r1 , k _r2x , ie with multi- _layered ring-shaped areas for the coding points around the point mark, the code word length can be increased ( FIG. 1e).

As a result of the entire identification code, the Dot mark image center point X₀, Y₀ and die Identification number passed on.

The image coordinate scaling factor of the image recording device can be applied directly to the contour coordinate calculation when determining the contour, ie the x coordinates of all determined contour points are divided by s, so that the subsequent consideration of the image coordinate scaling difference of the image recording device is omitted in all subsequent calculations. In this case, s _* X₀, Y₀ instead of X₀, Y₀ must be passed on as the result for the dot mark image center together with the identification number.

The entire image processing, validation and identification process for a point mark can be carried out after determining the ellipse parameters within a small image window around the center of the ellipse, the window size of which is determined using the determined ellipse _parameters and with constants k _r1.2 and k _u1.2 , so that all possible image structures as a point mark and possible image structures as associated coding points are within this image window. Through optimized image processing "within" this image window, the image structures are better determined and processed and thus the reliability and accuracy of the overall result is considerably increased. The pre-assigned angular positions for checking the concealment of the point mark by foreign objects should preferably only be assigned to the outermost coding points so that the concealment can be checked expediently.

The element obtained in the manner described above can then be used, for example, as an identification element in the General cargo sorting or as a target element at photogrammetric treatment of objects to be measured be used.

The coordinates and designations used in FIGS. 2a, b and c and in the description are listed separately below:

X, Y image coordinate system of the image recording device
x, y local image coordinate system on the center of the ellipse X₀, Y₀
x ′, y ′ coordinate system x, y rotated by the angle α
x ′ ′, y ′ ′ coordinate system x ′, y ′ after correcting the perspective distorted image
X₀, Y₀ image coordinates of the center of the ellipse in the coordinate system X, Y
a, b Half length of the major and minor axes of the ellipse
α Rotation angle of the ellipse
X _c , Y _c image coordinates of a coding point in the coordinate system X, Y
x _c ′ ′, y _c ′ ′ image coordinates of a coding point in the coordinate system x ′ ′, y ′ ′
θ ′, θ angular position of a coding point around the point mark in the coordinate system x ′, y ′ or x ′ ′, y ′ ′
r distance of a coding point from the center of the point mark in the coordinate system x ′ ′, y ′ ′
k _r1,2 , k _u1,2 constants of the _{search criteria} for coding points around the point mark.

For an exemplary explanation of the implementation of the method according to the invention, reference is made to the flow diagram attached as FIG. 3, from which an exemplary and preferred sequence of method steps results.

Claims (17)

1. Optoelectronically detectable identification or target element with a circular disk-shaped identification mark ( 2 ) delimited by a circular circumferential line, characterized by encoded circular disc-shaped coding elements ( 4 ) around the identification mark ( 2 ) on at least one concentric circular line. 2. Identification or target element claim 1, characterized in that the coding elements ( 4 ) are smaller than the identification mark ( 2 ). 3. Identification or target element according to claim 1 or 2, characterized in that the identification tag ( 2 ) in the identification tag center has an additional marking, such as a point or a cross. 4. A method for optoelectronic detection and processing of the coding of an identification or target element with an identification mark ( 2 ) having a circular circumferential structure and with coding elements ( 4 ) provided according to a ring code at certain positions with respect to the identification mark center, one having an identification or target element Object is imaged with a digital image recording device, comprising the following steps:

• - Determining the contour coordinates of all image structures of the image recognized on the basis of their contrast using computer-aided methods of image processing
• - Identification of image structures that are elliptical in the digital image
• - Determine the orientation of the elliptical image structures
• - Determining the actual arrangement of further structures in the elliptical structure of the digital image
• - Check whether the actual arrangement of further structures satisfies a predetermined rule or condition and accordingly eliminate or identify the elliptical structure as an identification mark ( 2 ) and further structures as coding elements ( 4 )
• - Determining the occupancy of the determined positions around the dog tag center by coding elements
• - Determining the code and an identification number of the identification or target element based on the actual arrangement of the coding elements.

5. The method according to claim 4, characterized in that an elliptical structure is identified as such by the ratio of Structure scope to structure area, the so-called Form factor, is calculated and such structures are eliminated, their form factor a predetermined Amount exceeds. 6. The method according to claim 5, characterized in that as a further distinguishing criterion minimal and Maximum values of a given structural scope are used and with the determined Structural scope can be compared. 7. The method according to claim 4, 5 or 6, characterized characterized in that when determining the orientation the elliptical image structures ellipse parameters are calculated and a coordinate transformation is performed so that in the digital image elliptical image structure is circular. 8. The method according to claim 7, characterized in that in a first transformation step a transformation into the main axis system of the ellipse and that in a second transformation step Rotation about a major axis is performed.   9. The method according to claim 7 or 8, characterized in that when determining the orientation of the elliptical Image structures the center coordinates X₀, Y₀, die Half length of the major and minor axes a, b and Angle of rotation can be calculated. 10. The method according to claim 9, characterized in that the ellipse parameters from an ellipse adjustment with Contour coordinates are calculated. 11. The method according to claim 9, characterized in that the ellipse parameters from the area moments of zeroth, first and second order can be calculated. 12. The method according to any one of claims 4 to 11, characterized characterized in that when determining the actual Arrangement of the other structures in the elliptical Structure or coding elements in the Dog tag calculates the center coordinates will. 13. The method according to claim 12, characterized in that the center coordinates of the coding elements a center of gravity calculation can be obtained. 14. The method according to any one of claims 4 to 13, characterized characterized in that in determining the actual  Arrangement of the coding elements their angular position θ and distance r from the dog tag center is calculated. 15. The method according to any one of claims 4 to 14, characterized in that for the identification of coding elements, the criteria according to equations and or be used. 16. The method according to any one of claims 4 to 15, characterized characterized in that the code information by Best fit and discretization of the angular position θ is obtained at predefined angular positions. 17. The method according to any one of claims 4 to 16, characterized characterized in that the center coordinates of the ellipse are taken as the desired target coordinates.