JP2011103032A - Optical automatic recognition code including three-dimensional shape areas - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To newly provide a height difference code allowing data to be carried and read by the arrangement of three kinds of heights, and in particular, allowing accurate cutout and reading even if there is noise around the height difference code or even if a code is inclined. <P>SOLUTION: In this optical automatic recognition code including a plurality of continuously arranged three-dimensional shape areas (unevenness), a common virtual reference plane in a height direction is set, and data are represented by the arrangement of the heights to the virtual reference plane of the three-dimensional shape areas. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an optical automatic recognition code and a reading method thereof. In particular, the present invention relates to an optical automatic recognition code that represents data by an arrangement of “height” with respect to a three-dimensional reference surface and a reading method thereof.

  Optical automatic recognition codes represented by bar codes are used in various ways as means for representing product information and the like. Examples of methods for marking these optical automatic recognition codes on an object include, for example, a method of sticking a seal printed with an optical automatic recognition code, and an optical automatic recognition code printed directly on an object using a printer or the like. The method of doing is known. The optical automatic recognition code attached in this way may cause the seal to peel off or cause ink to drop off in a marked environment (eg, in a freezer where frost or water droplets are likely to adhere). For example, data reading may be hindered.

  On the other hand, for such problems, the following prior art documents disclose a method of three-dimensionally forming a barcode using a mold having a concavo-convex pattern and a reading device thereof. According to this, it is possible to realize a three-dimensional barcode excellent in environmental resistance. However, the barcode disclosed here is merely a three-dimensional representation of a conventional barcode, and the amount of information can only hold the same amount of information as a conventional barcode. is there.

In addition, in order to read the three-dimensional barcode described in Patent Document 1, this uneven pattern is scanned with an optical sensor, and only the reflection from the surface of the convex portion is received and read in the same manner as a conventional barcode. 1 is disclosed.

  As described above, the three-dimensional barcode described in Patent Document 1 is a “three-dimensional” barcode, which represents a conventional barcode with a concavo-convex pattern. In addition, a surface having a certain size is required. For example, when this three-dimensional barcode is used in the above-described freezer, a certain smooth flat surface is required, and there is a possibility that restrictions may be imposed on articles and places that can be used.

JP 09-0669131

  As described above, marking and reading can be easily performed without being affected by peeling or bleeding of the marking portion even in a poor environment (for example, in a freezer where frost and water droplets are easily attached). No optical auto-recognition code is known so far.

  Therefore, the present invention applies the concept of the “color arrangement code” that has been proposed by the inventors of the present application, so that an optical automatic recognition code that carries and reads data in three kinds of height arrangement ( (Hereinafter referred to as “height difference code”). In addition, we propose a new optical automatic recognition code and its reading method that can be accurately cut out and read even when there is noise around the optical automatic recognition code or when the code part is tilted. To do.

(1) In the present invention, in order to achieve the above object, a common virtual reference plane in the height direction is set in the optical automatic recognition code composed of a plurality of three-dimensional shape regions arranged continuously, An optical automatic recognition code, wherein data is represented by an arrangement of heights of the three-dimensional shape region with respect to the virtual reference plane.

(2) Further, in the present invention, in the optical automatic recognition code including a plurality of three-dimensional shape regions arranged continuously, a common virtual reference plane in the height direction is set, and the three-dimensional shape region is An optical automatic recognition code characterized by having an inclined surface and representing data by an arrangement in the inclination direction of the inclined surface with respect to the arrangement direction of the three-dimensional shape region.

(3) In the optical automatic recognition code including a plurality of three-dimensional shape regions arranged continuously, the three-dimensional shape region has a surface having an inclination, and a common virtual reference plane in the height direction is An optical automatic recognition code that is set and represents data by an arrangement of a height of the three-dimensional shape region with respect to the virtual reference plane and an inclination direction of a surface having the inclination.

(4) Further, the present invention provides the optical automatic recognition code according to any one of (1) to (3), wherein the optical automatic recognition code has a substantial reference plane indicating shape indicating the virtual reference plane. This is an automatic expression recognition code.
(5) In the optical automatic recognition code according to any one of (1) to (3), the optical automatic recognition code according to the present invention has a cut-out reference shape indicating the three-dimensional shape region. It is a recognition code.

(6) In the optical automatic recognition code according to any one of (1) to (3), the present invention provides a substantive reference plane indicating shape indicating the virtual reference plane and a three-dimensional shape region. An optical automatic recognition code having a reference shape.

(7) The present invention is the optical automatic recognition code according to (1), wherein the three-dimensional shape region has three or more types of height. .
(8) Further, the present invention is the optical automatic recognition code according to (2), wherein the direction of the inclination of the surface of the three-dimensional shape region is four or more directions. is there.

(9) Moreover, the present invention provides the optical automatic recognition code according to (3), wherein the three-dimensional shape region has three or more types of heights, and the surface of the three-dimensional shape region is inclined. The optical automatic recognition code is characterized in that there are four or more directions.

(10) In the reading method of reading the optical automatic recognition code according to (4) and restoring the data, the present invention detects the reference plane indicating shape, and uses the virtual plane reference shape based on the reference plane indicating shape. An optical automatic recognition code reading method comprising a reference plane setting step for setting a plane.

(11) Further, in the optical automatic recognition code reading method for reading the optical automatic recognition code according to (5) and restoring the data, the present invention detects the cutting reference shape, and the cutting reference shape is Assuming that the three-dimensional shape region is present in the region to be shown, the three-dimensional shape present in the region indicated by the cutout reference shape is detected as the three-dimensional shape region representing the data. An optical automatic recognition code reading method comprising an area detection step.

(12) Moreover, this invention is an article | item which attached | subjected the optical automatic recognition code in any one of (1) to (9).

  As described above, the present invention is an “elevation difference code” that carries data in “height alignment” (elevation difference) with respect to a reference surface and a reading method thereof, and is affected by frost and water droplets such as in a freezer. Data can be read reliably even in an easy and inferior environment.

It is explanatory drawing of a height difference code. It is explanatory drawing at the time of seeing the elevation difference code in FIG. 1 from the direction perpendicular | vertical with respect to the arrangement | sequence direction of "the uneven | corrugated arrangement | sequence which forms a code | cord | chord". It is explanatory drawing of the object which has several planes. It is explanatory drawing at the time of making one height of "the uneven | corrugated arrangement | sequence which forms a code | cord | chord" of a height difference code the same as the height of a cut-out reference plane. FIG. 5 is an explanatory diagram when the height difference code in FIG. 4 is viewed from a direction perpendicular to the arrangement direction of the “concave and convex arrangement forming the code”. It is explanatory drawing when the uneven | corrugated arrangement | sequence which forms a code | cord | chord is located in the edge part of a target object. It is explanatory drawing at the time of providing a horizontal reference plane around the height difference code, and providing a cut-out reference plane around it. It is explanatory drawing at the time of using a "horizontal reference | standard index shape". It is explanatory drawing at the time of using "cutting standard shape". It is explanatory drawing of a slope code | cord | chord. It is explanatory drawing of a slope type height difference code.

Embodiment 1 (height difference code)
1. “Elevation code” “Elevation code” according to the present invention is an application of the concept of “color arrangement code” that has been proposed by the present inventors.

  The “color array code” is an optical recognition code that represents data only by the color array, and was originally developed by the inventor of the present application. These are all composed of three colored areas marked on the object and carry data depending only on the arrangement order of the color areas, so the shape and dimensions of the areas included in the code It has the feature of not depending on. According to this, even if the color arrangement code is deformed, the data is not affected and can be accurately decoded.

  However, if this color arrangement code is applied to a freezer where the influence of frost and water droplets is large, the color is easily affected by water droplets and dirt, making it difficult to detect the color due to fading or dirt. Will occur. Therefore, in order to apply in such an environment, it was desired to represent data using another parameter. Accordingly, the applicant of the present application has applied the concept of the color arrangement code to the invention of “level difference code” that represents data by “height arrangement” by examining the following.

  The color arrangement code carries data in the “color arrangement”, but in fact, if the plane (image) can be divided into at least three regions, basically the same concept (data carrying method) can be applied. In other words, if the areas are distinguished from 1, 2, and 3, a large amount of data is carried only by “the order in which the areas appear” as in “1-3-2-1-1-1-2-3-3”. Can be made. In the color arrangement code, for example, R, G, and B are set to 1, 2, and 3, so that the data is arranged in an arrangement such as "R-B-G-R-B-R-G-B-" To express.

  On the other hand, if there are two types of areas 1 and 2, the number of data that can be expressed is extremely small, such as “1-2-1-2-1-2-”. Therefore, when applying the data holding method of the color arrangement code, it can be said that it is preferable to divide the plane into at least three regions. It is preferable to select which element is used to distinguish the “region” according to the environment in which the optical automatic recognition code is used. Here, in consideration of a scene in which frost or water droplets are assumed in a freezer or the like, parameters that do not cause discoloration or fading are preferable. Therefore, the inventor of the present application first considered that it is suitable to distinguish the region by “three-dimensional shape”. Next, in order to classify three or more types, if three-dimensional shapes of “height” of low, medium, and high are continuously arranged and the data is expressed in that order, more data can be obtained. I came to realize that it would be possible to represent information. For example, in the above example, the regions 1, 2, and 3 are set to high, medium, and low, and data is represented in a sequence such as “high-low-medium-high-low-high-medium-low”. Such an optical automatic recognition code is referred to as an “elevation difference code”.

  Through such studies, the present inventor has realized, for the first time, a novel “height difference code” in which data is represented by a “three-dimensional shape” and “an arrangement of three or more heights”.

  Since this “height difference code” represents data by “height arrangement”, the width, area, shape, etc. of the unevenness do not affect the data. Because of this characteristic, it can be effectively used even in a poor environment such as the above-described freezer. In addition, since the invention described in the above-mentioned patent document 1 represents data with the width and shape of the unevenness as in the case of the conventional barcode, the degree of freedom of the shape of the unevenness is very limited. It has become.

  The inventor of the present application has also studied an imaging device used for reading. Since the color arrangement code needs to detect “color”, the image is picked up using a CCD camera or the like at the time of reading. On the other hand, the “height difference code” needs to detect the “height”, but it is somewhat difficult to accurately measure the “height” with a CCD camera or the like. Therefore, when reading the height difference code, it is preferable to use a three-dimensional measuring instrument that can measure the “height” of a three-dimensional shape.

  Although a technique similar to a three-dimensional measuring device is already known, here, as an example, a three-dimensional measuring device using the phase of infrared light is assumed as a “three-dimensional capture camera”. This technology receives the reflection from the location where each pixel of the area sensor corresponds to the reflection of the modulated infrared light, and detects the distance from the measuring device to the corresponding location based on the phase, and it is instantaneous This is a mechanism for detecting the three-dimensional shape of the region (distance from the measuring device at each position) (for example, Swissranger, etc. manufactured by Mesa, Switzerland http://www.j-clavis.co.jp/Mesa-sr/index. html). By applying this technology, the “height (height difference)” of a three-dimensional shape that could not be known by a CCD camera can be measured in an instant. As described above, the height difference code according to the present invention can be optically recognized, and thus can be said to be one of the optical automatic recognition codes.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described with reference to the drawings.

  FIG. 1 shows an example of a height difference code in which data is carried in three kinds of height arrays of “high”, “medium”, and “low”, and FIG. 2 shows a height difference code of FIG. Is shown in an explanatory view when viewed from a direction perpendicular to the arrangement direction of the “concave and convex arrangement forming the code”. The “height arrangement” means an arrangement (change) of heights of “high, medium, low”. Each concavo-convex pattern in this concavo-convex arrangement is a suitable example of the “three-dimensional shape region” in the claims, and the state in which each concavo-convex pattern holding the data is “continuously arranged in multiple numbers” is The “concave and convex arrangement forming the code”. That is, this “uneven arrangement for forming a code” is a preferred example of “a plurality of three-dimensionally arranged regions arranged continuously” in the claims. In addition, in the state of being “continuously” arranged, in addition to the state in which the three-dimensional shape regions are arranged without a gap, the state is arranged in a state of being considered “continuous” within a certain fixed interval. This includes cases where

  Here, since the “height” of the region can be distinguished by a relative height, assuming a certain reference surface, it is only necessary to distinguish low, medium, and high according to the height from the reference surface. . Therefore, assuming that the lowest part of the “concave and convex arrangement forming the code” is “virtually” as the concave and convex reference plane, it is referred to as “virtual concave and convex reference plane 10”. The “level difference code” represents data in a sequence of “height” with reference to the “virtual unevenness reference surface 10”. The “virtual unevenness reference plane” is a preferred example of the “virtual reference plane” in the claims.

  In FIG. 2, “a” represents “high”, “b” represents “medium”, “c” represents “low”, and “o” represents a surface around the height difference code. In FIG. 2, “c” corresponds to the lowest portion of the “concave / convex arrangement forming the code”, and is therefore the same as the virtual uneven reference surface 10. Thus, if the virtual uneven reference surface 10 is used as one of the “heights” (used as “low” here), the height difference code can be formed with fewer types of unevenness.

  In the height difference code, the data is represented by “height sequence”. For example, the sequence of “a → b”, “b → c”, “c → a” represents 1, and “a → c” “c → b”. Assuming that the sequence of “b → a” represents 0 (binary system), the height difference codes in FIGS. 1 and 2 represent “100001011011”. Although the binary system is explained here, it is also possible to carry data on the “height” itself instead of the “line of heights”, and in that case, the ternary is expressed by three types of “high”, “medium” and “low”. It can also represent legal data. In this case, it may be difficult to determine the boundary between data when the same value continues. In such a case, it is possible to easily detect the boundary of the data by determining the width of the unevenness and by determining in advance such as always returning to zero.

  As described above, the height difference code is an optical automatic recognition code that represents data by arranging the heights of a plurality of three-dimensional shape regions that are continuous (or only when they are continuous).

2. Cutting and reading of height difference code The present inventors have also proposed various methods for reading the color arrangement code described above (for example, Japanese Patent Application Nos. 2007-130504 and 2007-164286). ). A feature of these algorithms is that the color arrangement code can be efficiently cut out from the target image without depending on the shape and size of the region.

  On the other hand, as described above, the height difference code according to the present embodiment, as with the color arrangement code, data is represented only by the “height arrangement”, and the area (width) and shape of each region are It has the feature of not depending. Since it has such characteristics, data can be carried even in a poor environment where frost or water droplets adhere.

  Therefore, it is considered that it is preferable to construct the reading method by applying the above-described principle of reading the color arrangement code as it is to the height difference code according to the present embodiment. However, in the process of extracting and reading the height difference code from the captured image, there are the following problems specific to the height difference code.

Noise Problem The “level difference code” described so far can be realized by simply adding a level difference to a three-dimensional shape to form a “concave / convex arrangement forming a code”. However, various height patterns (noise patterns) other than the height difference code may exist within the imaging range of the imaging device such as a three-dimensional capture camera. That is, various objects other than the height difference code may exist in the screen (reading range) of the imaging apparatus, and in this case, each has a distance to the measuring instrument. These become noise, and various height patterns other than the normal height difference code are returned to the measuring instrument. In order to find the target concavo-convex arrangement (the concavo-convex arrangement that forms the code) and specify (cut out) the range from this, it is necessary not only to provide a height difference but also to make some contrivance.

Inclination problem Also, when the height difference code on the object is inclined, it is expected that an accurate reading will be hindered. Specifically, the target “uneven arrangement forming the code” is not necessarily directly facing the measuring device, and usually has some inclination. In such a case, the height at which the same value should be returned may vary depending on the position of the array. These can be calibrated by knowing the slope of the height difference code in advance, but if the height difference code itself cannot be cut out, the slope cannot be known after all. Furthermore, as described above, it is thought that it is more difficult to know the inclination in the presence of noise around the surroundings. For this reason, some device for accurate reading is required even when the height difference code is inclined.

  The inventor of the present application has made extensive studies on such problems, and succeeded in solving these problems by applying the following device to the “height difference code”.

3. Embodiments A method for solving the above-described problem regarding the height difference code described above, that is, a code carrying data at a height (change) will be described.

Example 1 Use of “Cutout Reference Plane” In this embodiment, a method for solving the above-described noise problem by providing “cutout reference plane 20” in a height difference code will be described. This “cut-out reference surface 20” is provided around (all around) the “corrugation array 40 forming the code” of the height difference code so as to be parallel to and flat with the “virtual unevenness reference surface 10”. Here, the “plane” refers to a region having an area larger than a certain reference value. This “plane” is not limited to a surface having no unevenness, and has an unevenness as long as it is an extremely small step (irregularity) compared to the step (unevenness) of the “corrugation arrangement 40 forming the code” of the height difference code. The surface may be. In this way, by providing the cut-out reference surface 20 on the “entire circumference” of the concave / convex array 40 that forms the code, the concave / convex pattern (noise pattern) that is not surrounded by a certain plane and the concave / convex array 40 that forms the code, It is possible to cut out with distinction. Further, since the periphery of the concavo-convex arrangement 40 forming the code is a flat surface having a certain width, the noise pattern can be eliminated in the vicinity of the height difference code, and accurate data can be read. Therefore, the above-described noise problem can be solved by providing the cut-out reference plane 20.

  The “cutout reference plane 20” is a suitable example of “a cutout reference shape indicating a three-dimensional shape region” in the claims.

  In this case, the height difference code can be cut out by the following algorithm.

Cutting / reading algorithm Cutting and reading of height difference codes will be described with reference to FIG. FIG. 3 shows an object having planes a to f. In this case, in the process of searching for the code with the data (three-dimensional high and low data) obtained with the three-dimensional capture camera,
(1) A “plane” having a certain spread is searched from the entire imaging data.

(2) Among the above “planes”, a plane having a concavo-convex pattern included in the plane is set as a “cutout reference plane candidate”. In FIG. 3, among the planes A to F, the plane D 1 having the uneven pattern is a “cutout reference plane candidate”.

(3) Each of the “cutout reference plane candidates” is converted into a rotational movement into a plane perpendicular to the optical axis of the three-dimensional capture camera by integrally converting the uneven pattern included therein,
(4) The height of the concavo-convex pattern to be included is quantized to a predetermined parameter.

(5) The quantized parameters are converted into data (decoded) in accordance with a predetermined rule to obtain original data.

(6) Naturally, planes that cannot be converted into data in the above process are excluded as “non-code ranges” and the planes including them are excluded.

In this algorithm, “a plane having a concavo-convex pattern” is cut out as a reference plane candidate, and the concavo-convex pattern included in the plane is cut out and read (see (3) above). Therefore, in the case of the “concave / convex pattern included in the plane”, even an unevenness (noise pattern) other than the “concave / convex arrangement forming the code” is cut out. However, even in such a case, those that cannot be quantized by the parameters given in advance are excluded as “non-code ranges” (see (4) and (5) above), and therefore are not affected by reading. rare. For example, if there is a recess that is too deep (too low) from a predetermined standard, it cannot be the target of quantization. For such a concavo-convex pattern, quantization fails in (4) above and is excluded from the reading target.

In addition, it is preferable that a format is defined in advance for the uneven pattern in order to represent predetermined data. For example, it is preferable to provide a format such as providing a certain pattern for the start bit and the end bit, and exclude those that are not based on the rules of the format from being decoded. In this way, more accurate reading can be performed by excluding so-called noise that is not a decoding target.
In FIGS. 1 and 2, the height difference code “corrugated arrangement 40 forming the code” and the “cutting reference surface 20” are formed at different heights. That is, the concavo-convex arrangement 40 forming the code of the height difference code is composed of three kinds of heights, a point, b point and c point. On the other hand, the cut-out reference plane 20 is positioned at the height o (see FIG. 2).

  On the other hand, as shown in FIGS. 4 and 5, one height (here, “a” representing “height”) of the “concave / convex arrangement 40 forming the code” is made equal to the height o of the cut-out reference plane 20. It is also suitable. In this case, the “concave / convex arrangement forming the code” is divided into several parts and included in the cut-out reference plane 20. Therefore, there is a possibility that the starting point and the ending point are unclear as to which range the “corrugation array 40 forming the code” of the height difference code is. Accordingly, it is preferable that predetermined conditions are determined in advance in the format of the height difference code so that the presence of the uneven arrangement 40 forming the code can be detected more reliably.

  As such conditions, the concave / convex array 40 forming the code is arranged linearly, and the concave / convex array 40 forming the code is a single concave / convex pattern in which the tip and end are determined. It is preferable to set conditions such as these.

  The cut-out reference plane 20 is provided on “the entire circumference” of the height difference code as described above. In other words, this state can also be said to “enclose” the “uneven arrangement 40 forming the code” on the cut-out reference surface 20. In other words, as described above, by making the concave / convex arrangement 40 forming the code on the cut-out reference plane 20 “include”, it becomes easy to distinguish the concave / convex (noise) other than the concave / convex representing the data. On the other hand, when the concavo-convex array 40 forming the code is not “included” in the cut-out reference plane 20, that is, when the concavo-convex array 40 forming the code is located at the end of the object (FIG. 6), there is a possibility that an uneven portion that becomes noise exists in the vicinity of the uneven portion array 40 forming the code, and it may be difficult to cut out due to the influence, or erroneous data may be read. Come. Therefore, as shown in FIG. 1 and FIG. 4 described above, the “corrugation array 40 forming the code” of the height difference code is formed at a position other than the end of the object so as to be “included” on the cut-out reference plane. It is preferable.

Embodiment 2 Addition of “Horizontal Reference Surface” In Embodiment 1 described above, by providing the “cutout reference plane 20” to the height difference code, it is easy to cut and read the height difference code even when noise exists. I explained that.

  In the present embodiment, an example in which a “horizontal reference plane 30” is further provided to solve the above-described “tilt problem” will be described. That is, by providing a “horizontal reference plane 30” in addition to the “cutout reference plane 20”, the tilt problem can be solved in addition to the above-described noise problem.

  Also in this case, it is preferable that the “concave / convex arrangement 40 for forming data” is formed with three or more different heights (for example, high, medium, and low). What is characteristic in this embodiment is that a “horizontal reference plane 30” is provided in addition to the “cutout reference plane 20”. Specifically, a horizontal “horizontal reference plane 30” parallel to the virtual irregularity reference plane 10 is provided around (around the entire circumference) of the “corrugation array 40 forming the code” of the height difference code, and the horizontal A flat “cutout reference surface 20” is provided around the reference surface 30. However, in the present embodiment, the cut-out reference plane 20 does not need to be parallel to the horizontal reference plane 30. FIG. 7 shows an example in which a horizontal reference plane 30 is provided around the height difference code, and a cut-out reference plane 20 is provided around the horizontal reference plane 30.

  The height difference code represents data in a sequence of “height” with reference to the virtual uneven reference surface 10. Accordingly, the inclination of the height difference code is the same as the inclination of the virtual uneven reference surface 10. In the present embodiment, since the horizontal reference surface 30 is provided “parallel” to the virtual uneven reference surface 10, the inclination of the virtual uneven reference surface 10 can be detected based on the horizontal reference surface 30. As a result, the inclination of the height difference code can be calibrated.

  This “horizontal reference plane 30” is a preferred example of “substantial reference plane indicating shape indicating a virtual reference plane” in the claims.

  In the first embodiment described above, the cut-out reference surface 20 is provided “in parallel” with the virtual concavo-convex reference surface 10, so that it also serves as a “horizontal reference surface”. That is, as shown in FIGS. 1 and 4, it is also preferable that the “cutout reference plane 20” and the “horizontal reference plane 30” be the same plane.

  Thus, in the present embodiment, since the horizontal reference plane 30 is also provided in addition to the cut-out reference plane 20, in addition to the presence of the noise pattern, even if the elevation difference code is inclined, the elevation difference code Cutting and reading can be performed smoothly. The algorithm of such operation will be sequentially described below.

Extraction / reading algorithm In this case, the process of searching for the height difference code using the data (three-dimensional height data) obtained by the three-dimensional capture camera is as follows.

(1) Search for “plane 1” having a certain spread from the entire imaging data.

(2) Check whether there is another “plane 2” included in the “plane 1”.

(3) Check whether or not there is an uneven pattern included in “Plane 2”.

(4) The planes 1 and 2 and the plane 2 on which the inclusion relation between the concave and convex patterns can be confirmed are set as horizontal reference plane candidates.

(5) The horizontal reference plane candidate is rotationally converted into a plane perpendicular to the optical axis of the “three-dimensional capture camera”,
(6) The height of the concavo-convex pattern to be included is quantized to a predetermined parameter.

(7) The quantized parameters are converted into data (decoded) according to a predetermined rule to obtain original data.

  In FIG. 7, the shape of the concavo-convex pattern of the height difference code is not “rectangular” as in FIGS. 1 and 4, but “ellipse shape”. This assumes the case of processing with an end mill or the like. As described above, the shape, size, arrangement direction of the uneven pattern of the elevation difference code, and the gap between the uneven patterns may be anything. Because the height difference code carries data only in “height”, the shape, dimensions, arrangement direction of the uneven pattern, gaps between the uneven patterns, etc. do not affect the data represented by the height difference code. It is because it has. This feature is the same as the “color arrangement code” already proposed by the present inventor. That is, since the “color array code” carries data only in the “color array”, it does not affect the data regardless of the shape or size in which the color is drawn. Similarly, the “height difference code” according to the present invention also carries data only in the “height array” with respect to the virtual concavo-convex reference surface 10, so the shape and dimensions of the concavo-convex representing the “height” are as follows: It does not directly affect the data.

  Here, we have described marking the difference code by drilling holes in the object with an end mill or the like. Of course, if the object can be molded (marked) at the manufacturing stage, such a method is used. May be used. In addition to the excavating means such as the end mill, it is also preferable to mark the height difference code by cutting the object with various etching means. Conversely, it is also preferable to express a difference in height by stacking predetermined members on the object to form a cord. In addition, any marking method may be used as long as the method can express the height difference on the object.

Example 3 Use of “Horizontal Reference Index Shape 35” The “horizontal reference plane 30” described so far is configured by a plane, but “shows the inclination of the virtual uneven reference plane in order to know the inclination of the height difference code. As long as the purpose of "" can be achieved, "horizontal reference index shape 35" that is a shape representing two or more line segments may be used. The “horizontal reference index shape 35” can be formed by, for example, two or more curved surfaces or ridge lines.

  In the third embodiment, two horizontal reference indicator shapes 35 a and 35 b are used to represent a line segment parallel to the horizontal reference plane 30. The horizontal reference indicator shapes 35a and 35b represent planes that are the horizontal reference plane 30 by representing different line segments. Here, line segments having two different directions are used. However, as described above, various shapes such as a curved surface and a ridge line can also be used.

  The “horizontal reference index shape 35” is a preferred example of the “substantial reference plane indicating shape indicating a virtual reference plane” in the claims.

  FIG. 8 shows an example when the “horizontal reference index shape 35” is used. In FIG. 8, the ridge lines 50 a and 50 b of the horizontal reference index shapes 35 a and 35 b are parallel to the virtual uneven reference surface 10 to represent the horizontal reference surface, thereby achieving the above object. Here, the ridge line means a line connecting the uppermost parts of the horizontal reference index shape.

Example 4 Use of “Cutout Reference Shape 25” The “cutout reference surface 20” has been described so far as being configured by a flat surface. If the purpose of “specifying the reference index shape 35” can be achieved, the “cutout reference shape 25” having a certain shape may be used. FIG. 9 shows an example in which “cutout reference shapes 25 a and 25 b” indicating “corrugated arrangement 40 forming a code” with a “triangle” shape are used. In this case, in the step of cutting out the height difference code, it is possible to distinguish from the noise pattern by searching for a triangle and taking only the concavo-convex pattern arranged between the triangles as a reading target.

  In FIG. 9, the heights of the surfaces of the cutout reference shapes 25a and 25b are formed in parallel with the virtual unevenness reference surface. In this way, the cutout reference shapes 25a and 25b can be used as the horizontal reference indicator 35. That is, in the cutout process, a triangle is searched for, and the height difference code can be read by correcting the inclination from this height.

  In FIG. 9, it is assumed that a plane parallel to the straight line connecting the cutout reference shapes 25 a and 25 b is parallel to the virtual uneven reference surface 10, and the height of the uneven array 40 that forms a code based on the virtual uneven reference surface 10. It reads the change.

  At this time, when the sizes of the cut-out reference shapes 25a and 25b are small, it may be difficult to specify the plane parallel to the line connecting them as the first place. In that case, a cutout reference shape 25c is newly provided, and a total of three cutout reference shapes 25a, 25b, and 25c are used to recognize that the line segment connecting them is parallel to the virtual unevenness reference surface 10. It is also preferable. Furthermore, when the cutout reference shape 25 is sufficiently large and has a somewhat large top area, the top surface of the cutout reference shape 25 directly represents the virtual uneven reference surface 10. Even in this case, in order to recognize the virtual concavo-convex reference plane 10 more accurately, a plurality of cutout reference shapes 25a and 25b are provided, and the virtual concavo-convex reference plane 10 is determined from the inclination of the flat portion of the top of the head. It is also preferable to recognize the inclination of the surface.

  In FIG. 9, the cut-out reference shape 25 indicates a concave / convex array 40 that forms a code, and at the same time, is a horizontal reference index 35 that corrects the inclination, and has two functions. However, it is also preferable to separate the two functions and provide the cut-out reference shape 25 and the horizontal reference surface 30 or the horizontal reference indicator shape 35 described above.

  The “cutout reference shape 25” is a preferred example of “a cutout reference shape indicating a three-dimensional shape region” in the claims.

  When the conditions for cutting out the height difference codes described so far have been rearranged, it is preferable to have the following two conditions when forming the height difference code.

A “Horizontal reference index” that is a shape that represents a “horizontal reference plane” that can indicate a “virtual unevenness reference plane” for tilt correction, or two or more line segments that can represent a “horizontal reference plane” It has “shape” (two or more curved surfaces, ridgelines, etc.) B “Cut-off reference surface” or “Cut-up uneven surface forming code” and “Horizontal reference surface” or “Horizontal reference index shape” Of course, the “cutout reference shape” exists. Of course, the B condition does not have to be satisfied in an environment where noise inclination cannot occur, and the A condition is satisfied in an environment where inclination problem cannot occur. It does not have to be. Furthermore, in an environment where none of the problems can occur, it is possible to accurately extract and read the height difference code even if none of the conditions of AB is satisfied (that is, the height difference code is simply formed with a simple uneven pattern). It can be carried out.

Embodiment 2 (supporting data of “surface tilt”)
In the first embodiment described above, the method for solving the noise problem and the inclination problem has been described with respect to the “level difference code” that carries data at a height (change). In the present embodiment, the “level difference code” concept is further applied, and the “slope code” in which data is carried in the “arrangement of the tilt direction of the surface” with the surface of each three-dimensional shape area as a slope. explain.

  The height difference code described so far is read by converting “continuity of height change” into data using a three-dimensional capture camera. If the inventor of this application uses this three-dimensional capture camera, it is possible to convert not the “array of heights” but the “array of tilt directions of the surface (array of tilt directions forming the code)” into data. Thought. In other words, since the 3D capture camera can detect the distance from the measuring instrument to the corresponding part, it detects not only the “height” of the object but also the “surface tilt direction” relative to the reference direction. It is also possible to do. Based on this idea, the inventor of the present application applied a “slope code” by applying the “level difference code” to the surface of the three-dimensional shape region as a slope, and supporting the “slope code”. It came to realize "." The “arrangement of the surface tilt direction” refers to an arrangement (change) of the surface tilt direction with respect to the array direction of the three-dimensional shape region.

  For example, FIG. 10B shows an example in which a “slope code” in which a three-dimensional shape region having a certain inclination (slope) is continuously arranged is formed. Here, as shown in FIG. 10 (1), the inclination direction of the surface is distinguished by four directions of “forward / reverse up / down” with reference to the arrangement direction. In this way, four values (that is, forward / reverse up / down) can be assigned in the direction of each inclination even if there is only one kind of inclination. To represent data in this sequence, tentatively, the sequence of “forward → reverse”, “reverse → up”, “up → down”, “down → forward”, “forward → up”, and “reverse → down” is 0, “forward → down” If the sequence of “lower → upper”, “upper → reverse”, “reverse → forward”, “upper / forward”, and “lower → reverse” is 1, the height difference code in FIG. 10 (2) represents “00001001”. It becomes.

  As described above, the “slope code” is an optical automatic recognition code that represents data in an array of the tilt directions of the surfaces of a plurality of three-dimensional shape regions that are continuous (or only if they are continuous).

  In this “slope code”, the “height” of each pattern (three-dimensional shape region) does not affect the data, so it may be formed at any height.

  In FIG. 10, in order to solve the problem of noise, a cut-out reference plane 120 is provided around the concavo-convex array 140 forming the code, as in the first embodiment. Of course, it is also preferable to use the cutout reference shape 125 instead of the cutout reference surface 120. Conversely, if noise problems cannot occur, these may not be formed.

  Further, in FIG. 10, the ridgeline 150 of each three-dimensional shape area forming the code is made parallel to the virtual unevenness reference surface 110 in order to solve the tilt problem. Thereby, the ridgeline 150 can be used instead of the horizontal reference indicator shape 135, and as a result, the slope of the slope-type height difference code can be detected.

  The ridge line is a line located in the vicinity of the uppermost part of each three-dimensional shape region, and is preferably a “line” having a length that can represent at least the inclination of the virtual unevenness reference surface. Therefore, in order to correct the tilt using the ridgeline 150, it is preferable that the uppermost part of the three-dimensional shape region is a “prism” so that a “line” is formed. In the case of “cylinder”, the vicinity of the top is a “point”, and it may be difficult to correct the inclination with the ridgeline. In such a case, it is preferable to provide the horizontal reference plane 130 and the horizontal reference indicator shape 135 separately.

  Here, the inclination correction is performed using the ridge line 150. Of course, it is also preferable to correct the inclination using the horizontal reference plane 130 and the horizontal reference indicator shape 135 as in the first embodiment. On the other hand, if the problem of tilt cannot occur, these need not be formed.

  The “slope code” has the same basic structure as the above-described “height difference code” except that the method of carrying data is “array in the direction of the surface inclination”. Therefore, this slope code can be cut out and read in the same manner as in the above-described height difference code (Embodiment 1).

Embodiment 3 (Data is represented by a sequence of “surface inclination and height”)
In the second embodiment above, an example is shown in which data is carried only for the "surface inclination", but in this embodiment, the data for the "surface inclination and height" is carried. An example will be described. Such an optical automatic recognition code is called a “slope-type height difference code”. If data is represented by both “tilt” and “height” as in this embodiment, more information can be represented without increasing the pattern (surface), and the information density can be increased. Become.

For example, if data is carried on three types of “high, medium, and low” as the height and four types of “front, back, up, and down” as the inclination, the data can be carried on the types of 12 levels on each surface. Can do. For example, in the case of an 11-plane configuration as shown in FIG. 11, if any one of 10 types of data from 0 to 9 is carried on each plane, approximately 10 ¹¹ different data can be represented by simple calculation. Is possible.

  In this way, the “slope-type height difference code” is an optical automatic recognition that represents data in a sequence of “surface inclination and height” of a plurality of three-dimensional shape regions that are continuous (or considered to be continuous). Code.

  In FIG. 11, only the concavo-convex arrangement forming the code is shown. However, as described above, in order to solve the noise problem and the inclination problem, a cut-out reference plane (or cut-out reference shape), It is also preferable to use a horizontal reference plane (or horizontal reference index shape).

  The basic structure of the “slope-type height difference code” in the present embodiment is the same as that of the “height difference code”, so that the cutout and reading are performed in the case of the height difference code described above (Embodiment 1). Can be done as well.

10, 110 Virtual reference surface 20, 120 Cutout reference surface 25, 125 Cutout reference shape 30, 130 Horizontal reference surface 35a, 35b, 135 Horizontal reference index shape 40, 140 Concave / convex arrangement representing data 50a, 50b, 150 Ridge o Height Height around the difference code a Height “High”
b Medium height
c Height “Low”
A plane B plane C plane D plane E plane F plane H height L inclination direction

Claims (12)

In the optical automatic recognition code consisting of a plurality of three-dimensional shape regions arranged continuously,
A common virtual reference plane in the height direction is set,
An optical automatic recognition code, wherein data is represented by an arrangement of heights of the three-dimensional shape region with respect to the virtual reference plane. In the optical automatic recognition code consisting of a plurality of three-dimensional shape regions arranged continuously,
A common virtual reference plane in the height direction is set,
The three-dimensional shape region has an inclined surface,
An optical automatic recognition code, characterized in that data is represented by an arrangement in an inclination direction of a surface having an inclination with respect to an arrangement direction of the three-dimensional shape region. In the optical automatic recognition code consisting of a plurality of three-dimensional shape regions arranged continuously,
The three-dimensional shape region has an inclined surface,
A common virtual reference plane in the height direction is set,
An optical automatic recognition code, wherein data is represented by an arrangement of a height of the three-dimensional shape region with respect to the virtual reference plane and an inclination direction of a surface having the inclination. In the optical automatic recognition code according to any one of claims 1 to 3,
An optical automatic recognition code having a substantive reference plane indicating shape indicating the virtual reference plane. In the optical automatic recognition code according to any one of claims 1 to 3,
An optical automatic recognition code having a cutout reference shape indicating the three-dimensional shape region. In the optical automatic recognition code according to claims 1 to 3,
A substantive reference plane indicating shape indicating the virtual reference plane;
A cut-out reference shape indicating the three-dimensional shape region;
An optical automatic recognition code comprising:
The optical automatic recognition code according to claim 1,
An optical automatic recognition code, wherein the three-dimensional shape region has three or more types of height. In the optical automatic recognition code according to claim 2,
An optical automatic recognition code characterized in that the direction of inclination of the surface of the three-dimensional shape region is four or more. In the optical automatic recognition code according to claim 3,
An optical automatic recognition code characterized in that there are three or more types of height of the three-dimensional shape region, and there are four or more directions of inclination of the surface of the three-dimensional shape region. In the reading method which reads the optical automatic recognition code according to claim 4, and restores the data,
An optical automatic recognition code reading method comprising: a reference plane setting step of detecting the reference plane indicating shape and setting the virtual reference plane from the reference plane indicating shape. In the optical automatic recognition code reading method of reading the optical automatic recognition code according to claim 5 and restoring the data,
The cutout reference shape is detected, and it is assumed that the three-dimensional shape region is in the region indicated by the cutout reference shape, and the three-dimensional shape present in the region indicated by the cutout reference shape represents the data. A three-dimensional shape area detecting step for detecting the three-dimensional shape area;
An optical automatic recognition code reading method comprising:   An article provided with the optical automatic recognition code according to any one of claims 1 to 9.

Images