CN101908156B - Image index structure - Google Patents

Abstract

The invention provides an image index structure, which comprises a content data part and a table head part. The content data part comprises a plurality of micro-image units, and the area occupied by the content data part is divided into a plurality of state areas; each state area is provided with a micro-image unit, and the micro-image unit is selectively positioned in one of a plurality of virtual areas formed by the equally divided state areas. The header includes a plurality of microimage units, and each microimage unit is arranged in a predetermined manner to provide header information identifying the image index. The invention can provide lower density of micro image unit; and an image index can be constructed with a small number of microimage elements.

Description

Image Index Structure

This application is a divisional application of an invention patent application with an application date of April 26, 2006, an application number of 200610075770.8, and an invention title of "Image Index Structure".

technical field

The invention relates to image recognition (pattern/image recognition) technology, specifically a data output and input method using an image index structure.

Background technique

FIG. 1 is a schematic diagram of a graphical indicator 102 formed on a surface 100 of an object. As shown in FIG. 1 , the image index 102 is composed of many micro image units. Since the micro image unit is quite small, it is easy to be ignored visually, or interpreted as the background color by human eyes. The image index 102 and main information 104 (such as the character pattern “APPLE” in FIG. 1 ) are jointly formed on the surface 100 of an object such as paper by means such as printing. The image index 102 corresponds to an index data, and does not affect the reception of the main information 104 by human eyes.

FIG. 2 is a schematic diagram of an electronic system 110 having an optical device 112 , a processing device 114 and an output device 116 for executing an image recognition program to read the image indicator 102 . The optical device 112 , the processing device 114 and the output device 116 are connected in a wired or wireless manner. The optical device 112 reads the surface of the object to obtain an enlarged image; then the processing device 114 takes out the image index 102 from the enlarged image and converts it into digital data, and obtains additional information corresponding to the digital data; finally, the output device 116 receives the additional information information, and output this additional information in a predetermined form. Therefore, through the design of the image indicator 102, more additional information can be carried on the surface of common objects such as book pages.

FIG. 3 is a schematic diagram of a conventional pattern design including a plurality of image indicators 102 . As shown in FIG. 3 , an image index 102 (the area surrounded by a dotted line) is formed by arranging a key point 202 , a plurality of grid points 204 and a plurality of information points 206 according to predetermined rules. Specifically, an image indicator 102 is centered on a key point 202 and a plurality of grid points 204 are arranged around it, wherein every four grid points 204 are configured as a rectangular block; and the center of every four grid points 204 is used as a virtual point , the information point 206 located in the rectangular box is selectively shifted to the upper, lower, left, and right sides of the virtual point by a certain distance, so as to represent different values respectively, and then read out by the above-mentioned electronic system 110 . The key point 202 is formed by shifting the grid point 204 at the center of an image index 102 to a predetermined direction for a certain distance. The key point 202 can provide a reference direction of the image index 102 as the optical device 112 reads the surface of the object to obtain An orientation reference when zooming in on an image. In addition, the arrangement of the four grid points 204 as a rectangular block can be used as a reference for correction if deformation occurs when the image index 102 is read or printed.

As shown in FIG. 1 , the image index 102 usually coexists with the main information 104 on the object surface 100; therefore, when the density of the micro-image unit is higher, the visual effect is worse, and the human eye is more likely to perceive the existence of the image index 102 and The chances of confusing image indicators 102 with primary information 104 increase. On the other hand, when the image index 102 is formed in a limited object surface area, under the premise of providing the same amount of information, the excessively high distribution density of micro-image units will cause the distance between two adjacent micro-image units to be too small; therefore , when the micro image unit is formed on the paper by printing, for example, it will cause relatively obvious visual interference. If this visual interference is to be reduced, it is usually necessary to further reduce the size of the micro-image unit, which will increase the resolution requirements of the printing press and paper, and it is also easy to miss dots during printing, which relatively increases the difficulty of printing and is likely to cause damage to the optical device. Interpretation errors or difficult to distinguish. The design method of the existing image index 102 shown in FIG. 3 obviously leads to an excessively high density of micro-image units, thereby causing various problems mentioned above.

Contents of the invention

The main purpose of the present invention is to provide an image index structure. It can provide lower micro-image unit density; and can form an image index with a small number of micro-image units.

The invention provides a data output and input method using an image index structure, characterized in that:

At least one image index structure corresponding to one index data is formed on the surface of an object, the image index structure includes a content data part and a header, the content data part includes a plurality of first micro image units and the content The area occupied by the data part is divided into a plurality of first state areas; each of the first state areas is provided with a first micro-image unit, and the first micro-image unit is selectively located in the plurality of areas formed by equally dividing the first state area. one of the virtual areas; the header includes a plurality of second micro-image units, and the second micro-image units are arranged in a predetermined manner to provide header information identifying the image index; The first state area and the second state area form a two-dimensional state area array, the second state area forms one row and one column in the two-dimensional state area array, and at least one of the second micro-unit images is biased shifting the center of the second state area to which it belongs, and the rest of the second micro-image units are arranged at the center of the second state area to which it belongs; and optically reading the surface of the object to obtain an enlarged image containing the image index structure , to retrieve the index data corresponding to the image index structure.

The header has 7 micro image units; the content data part has 9 micro image units and is divided into 9 state areas, and each state area is divided into first, second, third and third Four virtual areas: the micro image unit is selectively set in the first, second, third or fourth virtual area, so that the content data part has 262144 state combinations.

The 65536 state combinations in the 262144 states correspond to 65536 coding positions to which a plane belongs in the Unicode coding structure.

The header is distributed on the border of the content data part, and defines the micro image unit distribution area of the content data part.

The headers are distributed on two adjacent boundaries of the content data part to form an L-shaped distribution.

When two adjacent image indexes respectively correspond to different index data, the two adjacent image indexes have different headers, and the two different headers have different micro-image unit distribution modes.

The micro-image units are point-shaped or linear-shaped.

Each state area is equally divided into 8 virtual areas.

The micro image units of the virtual area are selectively placed near or far from the center of the state area to generate different state combinations.

The present invention also provides a data output and input method using an image index structure, characterized in that:

At least one image index structure corresponding to one index data is formed on the surface of an object, the image index structure includes a content data part and an L-shaped distribution header, the content data part includes a plurality of first micro image units and The area occupied by the content data part is divided into a plurality of first state areas, and each of the first state areas is provided with a first micro-image unit, and the first micro-image unit is selectively located in the equally divided first state area. One of the plurality of virtual areas formed; the L-shaped distribution table head includes a plurality of second micro-image units, and the second micro-image units are arranged in a predetermined manner to provide identification of the image indicators Header information, the first state area constitutes a two-dimensional array of state areas, at least one of the second micro image units is offset from the center of the second state area to which it belongs, and the rest of the second micro image units are set in the at the center of the second state area of ; and

The surface of the object is optically read to obtain an enlarged image containing the image index structure, so as to capture the index data corresponding to the image index structure.

The present invention also provides a data output and input method using an image index structure, characterized in that:

Forming a plurality of image index structures corresponding to an index data on the surface of an object, each of the image index structures includes a content data part and an L-shaped distribution header, and the content data part includes a plurality of first micro image units And the area occupied by the content data part is divided into a plurality of first state areas, and each of the first state areas is provided with a first micro-image unit, and the first micro-image unit is selectively located in the equally divided first state area One of the plurality of virtual areas formed, the L-shaped distribution header includes a plurality of second micro-image units, and the second micro-image units are arranged in a predetermined manner to provide identification of the image indicators The header information of the first state area constitutes a two-dimensional state area array, at least one of the second micro image units is offset from the center of the second state area to which it belongs, and the rest of the second micro image units are set in the At the central position of the second state area, and each of the first state areas is surrounded by three adjacent L-shaped distribution table heads; and

The surface of the object is optically read to obtain an enlarged image containing the image index structure, so as to capture the index data corresponding to the image index structure.

The invention provides a data output and input method using an image index structure, characterized in that:

At least one image index structure corresponding to one index data is formed on the surface of an object, the image index structure includes a content data part and a header, the content data part includes a plurality of first micro image units and the content The area occupied by the data part is divided into a plurality of first state areas; wherein each of the first state areas is provided with a first micro image unit, and the first micro image unit is selectively located in the area formed by dividing the first state area one of the plurality of virtual areas; the header includes a plurality of second micro image units, and the second micro image units are arranged in a predetermined manner to provide header information identifying the image index, The first state area constitutes a two-dimensional state area array, at least one of the second micro-image units is offset from the center of the second state area to which it belongs, and each of the first state areas is surrounded by a plurality of adjacent headers surrounding; and

The surface of the object is optically read to obtain an enlarged image containing the image index structure, so as to capture the index data corresponding to the image index structure.

The beneficial effects of the present invention are that a large amount of information is provided by using less number of dots (lower dot density), thereby providing a better visual effect, and it is less likely for human eyes to perceive the existence of image indicators without making the image Indicators are confused with co-existing primary messages.

Description of drawings

Fig. 1 is a schematic diagram of an image index formed on the surface of an object;

2 is a schematic diagram of an electronic system for reading image indicators;

Fig. 3 is a schematic diagram of an existing pattern design including multiple image indicators;

4 is a schematic diagram of a pattern formed by arranging a plurality of image indicators designed in an embodiment of the present invention;

FIG. 5 is an enlarged schematic diagram of an image index in FIG. 4 .

Fig. 6 is a schematic diagram of a micro-image unit's arrangement in the status area;

FIG. 7 is a schematic diagram of a bit array corresponding to the image index of FIG. 5 .

FIG. 8 is a schematic diagram of the same header design respectively corresponding to the same image index.

FIG. 9 is a schematic diagram of different header designs respectively corresponding to two adjacent different image indexes.

FIG. 10 is a schematic diagram of different header designs respectively corresponding to two adjacent different image indexes.

11A and 11B are schematic diagrams comparing the present invention with existing designs.

12A and 12B are another schematic diagram comparing the present invention with existing designs.

Fig. 13 is a schematic diagram of another embodiment of the image index structure of the present invention.

Fig. 14 is a schematic diagram of another embodiment of the image index structure of the present invention.

Fig. 15 is a schematic diagram of another embodiment of the image index structure of the present invention.

Fig. 16 is a schematic diagram of another embodiment of the image index structure of the present invention.

Detailed ways

FIG. 4 is a schematic diagram of a pattern formed by arranging a plurality of image indicators 10 according to an embodiment of the present invention, and FIG. 5 is an enlarged schematic diagram of one of the image indicators 10 to clearly illustrate the design of the present invention. As shown in FIG. 5 , the image pointer 10 includes a content data part 12 and a header 14 . According to this embodiment, the content data portion 12 includes 9 micro-image units composed of 9 dots 16; and the area occupied by the content data portion 12 is divided into 9 state areas 18, thereby forming a 3×3 plane Dimensional status area array such that each status area 18 includes a dot 16. According to the design of this embodiment, different placement positions of a dot 16 in a status area 18 can be used to represent a value in the corresponding index data. Specifically, as shown in FIG. 6, a status area 18 can be divided into four virtual areas, and dots 16 are selectively placed in the lower right, lower left, upper left, or upper right virtual areas to represent four different virtual areas. Bit value 00, 01, 10 or 11. Therefore, the arrangement relationship of the dots 16 shown in the content data portion 12 in FIG. 5 can correspond to the bit array shown in FIG. 7 . When the content data part 12 has 9 dots 16 and is placed in 9 state areas respectively, because each of the state areas 18 is divided into four virtual areas, when each dot 16 is selectively arranged in the four virtual areas One of them, the content data section 12 can have 4 ⁹ (=262144) state combinations. Therefore, by using the design of the image index 10 shown in FIG. 5 , 262144 different values in the corresponding index data can be represented. Therefore, according to the design of the content data part 12 of the present embodiment, 65536 state combinations can be taken out from the 262144 state combinations, so as to correspond to the 65536 encoding positions of a literal in the Unicode encoding structure; the remaining state combinations can be Reserved for other purposes, for example, state combinations may be provided as corresponding checksum code encoding locations.

On the other hand, since the image index 10 is composed of a group of micro image units, a header 14 needs to be provided to distinguish and isolate two adjacent image indexes 10 . As shown in FIG. 8 , the four image pointers 10 all have the same content data part 12 , that is, the same pointer data content, so the four image pointers 10 are all provided with the same specific header 14 . In this way, the same image index 10 can be found only by finding the specific header 14 without being interfered by other adjacent image indexes 10 . The header 14 also includes a plurality of micro-image units such as dots 16 , and the area occupied by the header 14 is divided into a plurality of status areas 18 . Referring to Fig. 5 again, according to the design of the present embodiment, each state region 18 of the meter head 14 includes a dot located at the center of the state region 18, so the meter head 14 has 7 dots; Surrounding the content data portion 12 with 9 dots on the adjacent boundary, the header 14 is formed into an L-shaped distribution, and the entire image index 10 is arranged in a 4×4 square matrix composed of 16 dots. As shown in Figure 5, the dot placement positions of the meter head 14 are all preset as the central position of a state area, thereby making the process of identifying the meter head 14 of the image index more rapid; but the meter head 14 wherein The position of one dot 16 ′ needs to be shifted in a specific direction different from the positions of other dots 16 . In this way, when image recognition is performed to read the image index 10, after the optical device (not shown) reads the surface of the object and obtains an enlarged image, as long as the header 14 of the image index 10 is recognized first, the image index 10 can be identified. The image pointer 10 is oriented to accurately capture the state combination of the content data portion 12 .

In addition, by changing the position of the dots 16 in the header 14, different headers 14 that are different from each other can be produced. Therefore, when each image index 10 corresponds to a different numerical value by adjusting the state combination of the content data part 12 , the image index 10 respectively corresponding to two different numerical values can distinguish the two through different headers 14 . As shown in FIG. 9 , two different headers 14 a and 14 b with different positions of the dots 16 can be used to distinguish the upper and lower adjacent data portions 12 a and 12 b with different contents. Alternatively, as shown in FIG. 10 , two different headers 14 c and 14 d with different positions of dots 16 can be used to distinguish left and right adjacent data portions 12 c and 12 d with different contents.

In addition, according to the design of this embodiment, the header 14 is formed on two adjacent borders of the content data portion 12 and defines the distribution range of the dots 16 of the content data portion 12 . Therefore, after the image is read by the optical device, if the image index 10 is distorted such as deformation, the header 14 can be used to correct the dot arrangement in the content data portion 12 accurately.

The advantages of the present invention over existing designs are illustrated below by means of comparative figures.

FIG. 11A is a schematic diagram of a prior art design, and FIG. 11B is a schematic diagram of a design of the present invention. Before explaining the advantages of the present invention, at first define an effective information ratio value:

E=(the number of dots representing the content information of the image index)/(the number of all dots).

As shown in Figure 11A, in a 5 * 5 dot matrix block of existing design, each information dot 206 is all surrounded by four grid dots 204; Therefore, an image index can be divided into a plurality of dotted lines Grid point pairs 22, each grid point pair 22 includes a grid point 204 and an information point 206; so the effective information ratio value of the existing design is 50%, and the value will not change with the block size of the dot matrix a fixed constant of . On the other hand, as shown in FIG. 11B, in terms of the present invention design of the same 5×5 dot matrix block, the effective information ratio value is (4×4) after deducting the header dots occupying two boundaries. )/(5×5)=64%. And according to the design of the present invention, when the image index 10 is larger, the effective information ratio value is higher. For example, a 10×10 dot matrix block has an effective information ratio of (9×9)/(10×10)=81%. It can be seen that the effective information ratio value of the present invention is obviously higher than that of the existing design, and can be further improved with the increase of the image index. In other words, the present invention can use fewer dots (lower dot density) to represent the same amount of information.

As far as the design of an image index is concerned, the number of micro image units should be reduced as much as possible according to the size of the micro image units and the distance between them, so that the influence of the image index on the overall brightness of the object surface is reduced. As mentioned earlier, the image index usually coexists with the main information on the surface of the object. Therefore, when the density of the dots is higher, not only the visual effect will be worse, but also the human eye is more likely to perceive the existence of the image index and make the image index and the main information Increased chance of confusion. Therefore, the effective information ratio value of the present invention is obviously higher than that of the existing design, so the same amount of information can be expressed with fewer dot numbers (lower dot density). This can provide a better visual effect and the human eyes are less likely to perceive the existence of the image index, so as not to confuse the image index with the main information. On the other hand, when the image index is formed in a limited object surface area, under the premise of providing the same amount of information, as shown in Figure 11A, the too high distribution density of dots leads to too small distance between two adjacent dots, Therefore, relatively obvious visual interference is generated, or it is easy to increase the difficulty of printing and cause misinterpretation or difficult identification of the optical device. The lower density design of the present invention does not cause this problem.

12A and 12B are another schematic diagram comparing the present invention with existing designs. As shown in Figure 12A, in an image index, at least one key point 202 needs to be formed in the existing design, so together with the four rectangular squares and information points 206 formed by four grid points 204 around it, at least 13 dots are required Only then can an image index be formed. However, as shown in FIG. 12B , according to the design of the present invention, only four dots 16 are needed to form an image indicator. Therefore, the present invention can form an image index with fewer dots, so that the arrangement on different object surfaces will be more flexible, and it is not easy to cause excessive dot distribution density.

FIG. 13 is another design example of the image indicator 10 of the present invention. The distribution of the dots 16 on a state area 18 of the content data portion 12 is not limited, as shown in FIG. 13 . When the state area 18 is divided into four virtual areas, it can be divided into two groups of dot distribution methods according to the distance from the center point P of the state area; the dots 16a that are closer to the center point P are set at the lower right, lower left, and upper left The square or the upper right virtual area can represent four different bit values 000, 001, 010 or 011 respectively; the dot 16b farther away from the center point P is set in the lower right, lower left, upper left or upper right virtual area, which can be represent four other different bit values 100, 101, 110 or 111, respectively. Through this design, the distribution of dots 16 on a state area 18 can produce 8 possible state combinations.

Of course, according to the design of the present invention, it is only necessary to place the dot 16 in one of the plurality of virtual areas to represent different bit values, and the number of virtual areas equally divided by each state area 18 is not limited. As shown in FIG. 14 , a state area 18 can also be divided into eight virtual areas, and each dot 16 can be selectively placed in one of the virtual areas to generate eight possible state combinations. In other words, the positions of the dots 16 in the present invention are not limited, and the status area 18 only needs to be divided into several virtual areas.

In addition, the micro-image unit of the image index 10 is not limited to the dot 16 shown in the above-mentioned embodiments, and may also be, for example, a dot-like object with other shapes. Of course, it is only necessary to achieve the purpose of identifying different states. The micro image unit is not limited to use a certain specific way of expression, for example, it can also be represented by a short line segment 24 as shown in FIG. 15 . In addition, the total number and arrangement of the micro-image units of the image index 10 are not limited, and the state area square matrix of the image index 10 is not limited to adopting a square matrix arrangement with equal length and width. For example, as shown in FIG. Equal rectangular square matrices are also available.

FIG. 16 is a schematic diagram of another design of the header 14 in an image pointer 10 . The header 14 is not limited to be distributed on the boundary of an image index 10 , and can also be arranged at the central part of the image index 10 as shown in FIG.

The above-mentioned embodiments are only used to illustrate the present invention, but not to limit the present invention.

Claims (13)

1. A data output and input method using an image index structure, characterized in that: At least one image index structure corresponding to one index data is formed on the surface of an object, the image index structure includes a content data part and a header, the content data part includes a plurality of first micro image units and the content The area occupied by the data part is divided into a plurality of first state areas, and each of the first state areas is provided with a first micro-image unit, and the first micro-image unit is selectively located in the plurality of areas formed by equally dividing the first state area. One of the two virtual areas, the header includes a plurality of second micro-image units and the area occupied by the header is divided into a plurality of second state areas, and the second micro-image units are configured in a predetermined manner Arranged to provide header information identifying the image index structure, the first state area and the second state area form a two-dimensional state area array, and the second state area forms the two-dimensional state area array In one row and one column, at least one of the second micro image units is offset from the center of the second state area to which it belongs, and the rest of the second micro image units are all arranged at the center of the second state area to which they belong; and The surface of the object is optically read to obtain an enlarged image containing the image index structure, so as to capture the index data corresponding to the image index structure. 2. The data output and input method using image index structure according to claim 1, characterized in that, each of the first state areas is equally divided into first, second, third and fourth virtual areas; and the The first micro-image unit is selectively arranged in the first, second, third or fourth virtual area to represent a value in the index data. 3. The data output and input method using image index structure according to claim 1, characterized in that, the header has 7 micro-image units; the content data portion has 9 first micro-image units and distinguishes There are 9 first state areas, each of which is divided into first, second, third and fourth virtual areas; the first micro image unit is selectively arranged in the first, second , The third or fourth virtual area, so that the content data part has 262144 kinds of state combinations. 4. The data output and input method using image index structure according to claim 3, characterized in that 65536 state combinations in the 262144 states correspond to 65536 encoding positions of a literal in the Universal Code encoding structure. 5. The data output and input method using image index structure according to claim 1, characterized in that, the header is distributed on the boundary of the content data part, and defines the micro image unit distribution of the content data part area. 6 . The data output and input method using the image index structure according to claim 5 , wherein the headers are distributed on two adjacent boundaries of the content data portion to form an L-shaped distribution. 7 . 7. the data output and input method using image index structure according to claim 1, is characterized in that, when two adjacent image index structures correspond to different index data respectively, described two adjacent image index structures have corresponding different headers, and the two different headers have different micro-image unit distribution modes. 8 . The data output and input method using an image index structure according to claim 1 , wherein the first and second micro image units are in the shape of points or lines. 9. The data output and input method using an image index structure according to claim 1, wherein each of the first state areas is equally divided into 8 virtual areas. 10. The data output and input method using image index structure according to claim 1, characterized in that, the micro-image units of the virtual area are selectively placed near or far from the central position of the state area to generate different state combination. 11. A data output and input method using an image index structure, characterized in that: At least one image index structure corresponding to one index data is formed on the surface of an object, the image index structure includes a content data part and an L-shaped distribution header, the content data part includes a plurality of first micro image units and The area occupied by the content data part is divided into a plurality of first state areas, and each of the first state areas is provided with a first micro-image unit, and the first micro-image unit is selectively located in the equally divided first state area. One of the plurality of virtual areas formed, the L-shaped distribution table head includes a plurality of second micro image units and the area occupied by the L-shaped distribution table head is divided into a plurality of second state areas, the The second micro-image units are arranged in a predetermined manner to provide header information identifying the image index structure, the first state area forms a two-dimensional array of state areas, at least one of the second micro-image units is offset The center of the second state area to which it belongs, and the rest of the second micro image units are set at the center of the second state area to which it belongs; and The surface of the object is optically read to obtain an enlarged image containing the image index structure, so as to capture the index data corresponding to the image index structure. 12. A data output and input method using an image index structure, characterized in that: Forming a plurality of image index structures corresponding to an index data on the surface of an object, each of the image index structures includes a content data part and an L-shaped distribution header, and the content data part includes a plurality of first micro image units And the area occupied by the content data part is divided into a plurality of first state areas, and each of the first state areas is provided with a first micro-image unit, and the first micro-image unit is selectively located in the equally divided first state area One of the plurality of virtual areas formed, the L-shaped distribution table head includes a plurality of second micro-image units and the area occupied by the L-shaped distribution table head is divided into a plurality of second state areas, so The second micro-image units are arranged in a predetermined manner to provide header information identifying the image index structure, the first state area forms a two-dimensional state area array, at least one of the second micro-image units is Move the center of the second state area to which it belongs, and the rest of the second micro-image units are set at the center of the second state area to which it belongs, and each of the first state areas is surrounded by three adjacent L-shaped distribution headers surround; and The surface of the object is optically read to obtain an enlarged image containing the image index structure, so as to capture the index data corresponding to the image index structure. 13. A data output and input method using an image index structure, characterized in that: At least one image index structure corresponding to one index data is formed on the surface of an object, the image index structure includes a content data part and a header, the content data part includes a plurality of first micro image units and the content The area occupied by the data part is divided into a plurality of first state areas; wherein each of the first state areas is provided with a first micro image unit, and the first micro image unit is selectively located in the area formed by dividing the first state area One of the plurality of virtual areas; the header includes a plurality of second micro-image units and the area occupied by the header is divided into a plurality of second state areas, and the second micro-image units are represented by a Arranged in a predetermined manner to provide header information identifying the image index structure, the first state area constitutes a two-dimensional array of state areas, at least one of the second micro-image units is offset from the center of the second state area to which it belongs , and each of the first state areas is surrounded by a plurality of adjacent table headers; and The surface of the object is optically read to obtain an enlarged image containing the image index structure, so as to capture the index data corresponding to the image index structure.