JP2018501588A - Four-dimensional code, image identification system and image identification method based on four-dimensional code, four-dimensional code retrieval system, and four-dimensional code retrieval method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a rapid and accurate search method by creating a four-dimensional code concept using image identification technology and setting a four-dimensional code standard and a network platform. A true color image, a two-dimensional code, a color superimposed on the two-dimensional code, an ID number, and a complete format including a plurality of correction points, and a true color image in which one correction point is installed in each square. To store an identification diagram including at least a half format including a pure image format including only a true color image and a series of identification data corresponding to the identification diagram, a receiving / shipping module, an identification data generating module, and identification data The identification map is obtained by the mobile terminal and sent to the server by the network, and the server receives the identification chart and analyzes the corresponding identification data. And contrasting with the identification data stored in the database, Provided is an image identification system characterized in that identification information is returned to a mobile terminal or a computer. [Selection] Figure 4

Description

  The present invention relates to the field of image identification, and more particularly to an image identification system, a four-dimensional code using the system, a search system therefor, and a search method.

  Currently, image identification technology is an important field studied by computers, and in recent years it has been rapidly developed and widely applied to almost every field, such as handwriting input, postal code identification, kanji identification, face identification, fingerprint identification and With iris maturity being relatively mature, these technologies are widely applied in daily life and have a significant impact on economics, military, national security and people's daily life.

  The current image identification technology is operated only in a few specific fields, and these operations cannot be immediately followed, and a wide range of operations in the universal field becomes more insufficient. There is no identification standard, specifically children's education books can convert image identification to audio reproduction without defining corresponding graphic identification standards and server database support. If you cannot express the knowledge of geometry, physics, chemistry, etc. of junior high school with characters during the network search due to its graphic structure or special codes, you can not search, and if you pay with a smartphone, you can check only the SMS number, If lost, it will be paid maliciously and a loss will occur, and palm palmprints will not be identified (handprints will be handy on smartphones), which should define corresponding graphics identification criteria and server database support, And it is necessary to operate software and network system support.

  The present invention creates a four-dimensional code concept using image identification technology, and provides a rapid and accurate search method by setting a four-dimensional code standard and a network platform.

  An object of the present invention is to provide an image identification system in order to solve the problem of the conventional technology,

True color image (true image in Chinese: color image with 24-bit or 32-bit color information per pixel), 2D code, color superimposed on 2D code, ID number, and multiple correction points An identification diagram including at least a complete format, a half format including a true color image in which one correction point is set in each square, and a pure image format including only a true color image, and a series of identifications corresponding thereto Data and
A server including a receiving / shipping module, an identification data generation module, a database for storing identification data, and a search determination module;

The identification diagram is obtained by the mobile terminal and sent to the server by the network, the server receives the identification diagram, analyzes the corresponding identification data, and compares it with the identification data stored in the database. Information is returned to the mobile terminal or computer.
The present invention
Procedure 1 for capturing identification map information by photographing with a mobile terminal, and sending it to a server via a network;

  An identification including a procedure 2 in which the server receives the identification diagram information, analyzes the corresponding identification data, and returns the identification information to the mobile terminal after being compared with the identification data stored in the server database. Provide a method.

The present invention provides a four-dimensional code, and includes an identification diagram and a series of identification data corresponding to the identification diagram. The identification diagram includes a true color image, a two-dimensional code, a color to be superimposed on the two-dimensional code, and an ID number. The true color image, the two-dimensional code, the color superimposed on the two-dimensional code, and the ID number having the same or corresponding index.
The present invention

A true color image each having a similar or corresponding index, a two-dimensional code, an identification diagram including colors and ID numbers superimposed on the two-dimensional code, and a series of identification data corresponding to the identification diagram;
A server including a receiving / shipping module, an identification data generation module, a database for storing identification data, and a search determination module;

After the identification map is acquired by the mobile terminal and sent to the server by the network, the server receives the identification map, analyzes the corresponding identification data, and compares it with the identification data stored in the database. A four-dimensional code retrieval system is provided for returning the message to a mobile terminal or computer.
The present invention
Procedure 1 for photographing with a mobile terminal to obtain four-dimensional code information and sending it to a server via a network;

  4 including a step 2 in which the server receives the four-dimensional code information, analyzes the corresponding identification data, and searches and compares the identification data stored in the database of the server, and then returns the identification information to the mobile terminal. A dimension code search method is provided.

  Compared with the prior art, the present invention has the following advantages.

The present invention creates a four-dimensional code concept using image identification technology, and provides a rapid and accurate search method by setting a four-dimensional code standard and a network platform.
The amount of data included in the four-dimensional code is large, and the encryption, identification and anti-counterfeit functions are strong.

  The present invention adopts a true color image as a data carrier, has an intuitive / affinity feature, allows a user to find a point of interest among many disturbing objects, and allows a designer to publish publications such as advertisements, videos, etc. In the case of designing, the identification diagram and the publication are well coordinated, and the user can immediately scan and photograph using equipment such as a smartphone, immediately acquire data, and immediately enter the related platform.

FIG. 4 is a complete format identification diagram of the present invention. It is a half format identification figure of this invention. It is a pure image format identification diagram of the present invention. It is an operation | movement principle figure of a figure, an image identification system, and a search system. It is a hierarchical structure figure of a database. It is a call flowchart of a smart phone. It is a flowchart of a four-dimensional code search method.

  Hereinafter, the present invention will be described in detail with reference to the specification drawings and specific examples.

As shown in FIGS. 1 to 3, the present invention provides an identification diagram and a four-dimensional code including a series of identification data corresponding to the identification diagram, the identification diagram being a true color image, a two-dimensional code, and a two-dimensional code. The color and ID number to be superimposed on each other, the true color image described above, the two-dimensional code, the color superimposed on the two-dimensional code, and the ID number have the same or corresponding index.
In this embodiment, the identification diagram includes three formats: full format, half format, or pure image format.

  The complete format contains a true color image, a two-dimensional code, a color and ID number superimposed on the two-dimensional code, the two-dimensional code is placed above the true color image, correction points are placed in the identification diagram, and the ID number Is placed below the true color image.

  A full-format true color image is placed between the identification diagrams, correction points are placed in each square of the identification diagram, a two-dimensional code is placed between the correction points on the left and right sides above the identification diagram, and the color is on the two-dimensional code. The ID number is set below the identification diagram and is located between the correction points on the left and right sides.

  It can be applied to special operating environments where the ID number is not scanned or cannot be communicated with a smartphone over long distances. By manually entering the ID number, data of other dimensions can be searched in the same way. Specify the corresponding complete information. The true color image in the true color image is a company trademark, company identification chart, product identification chart, public sign identification chart, etc., or a personal photograph, a personal identification chart, or the like. At the same time, an identification code is obtained by calculating the image, and the ID number and the code contained in the two-dimensional code can be searched interactively or independently (ie, the following pure image format), and the material can be retrieved from the database. Make it convenient.

  To display the data in an image format, the two-dimensional code is further coated with color, and a computer or smartphone terminal program conveniently acquires the image from the scanned image and converts it to data, which requires net marking. It can be completed directly without it. The data content included in the two-dimensional code and the color above it is the same as the ID number at the bottom of the four-dimensional code in the format, the ID number that is the total index of the four-dimensional code, and the operation of the four-dimensional code And a category number for classifying a category, a true color image category, a modified version, a special type, and the like. Among them, the operation classification may include corporate operation, safe operation, public operation, social operation, national standard, personal operation, multimedia operation, educational magazine, other operation, and the like. True color image classification may include common identification diagrams, high-resolution identification diagrams, facial shape identification diagrams, car number identification diagrams, palmprint identification diagrams, encryption password identification diagrams, product identification diagrams, educational book publication identification diagrams, etc. . The two-dimensional code and the colors covered by the two-dimensional code may further include contents such as company business cards, personal business cards, and product content descriptions.

There are eight correction points in the format, each is arranged in a square, the coloring points are distributed at the top, adopt a black background instead of a white background, and enhance the camera software border and brightness in smartphones, PCs and cameras With the compensation function, the correction point area is reduced, the correction point located below the four-dimensional code is a black and white point, and the white point has a black side. It is for strengthening.
Using the correction has the following advantages:

1) The reason for correcting the size is that when shooting, the image distance is caused by the shooting distance and angle, the lens quality, the object to be imaged, and the image is deformed. When the amount of image deformation increases, the abandonment by error is submitted. There are five image average values: Adjacent (hard edge retention), bilinear, bicubic (applied to smooth gradual change), bicubic comparative smoothing (applied to enlargement), and bicubic roller smoothing (applied to reduction). Adopt one mode.
2) Image alignment: The identification ability is improved by matching the four correction points in the terminal program with the four correction points of the square of the imaged object.
3) Color correction: The RGB values of the eight correction points are as follows:
Red point: RGB255, 0, 0 Light red point: RGB255, 127, 127
Green point: RGB0, 255, 0 Light green point: RGB127, 255, 127
Blue point: RGB0, 0, 255 Light blue point: RGB127, 127, 255
White points: RGB255, 255, 255 Black points: RGB0, 0, 0

  The half format includes a true color image and a plurality of correction points placed in the true color image square. Half format is used in special environments, for example, for identification of educational books, using this format for children's reading, taking a reading with a smartphone, then the smartphone emits a sound, eg what is this to the child, or This figure tells whether to read or sing a story with this figure, or by shooting a geometrical subject in a textbook of a junior high school with a smartphone, it displays a solution concept or answer on the smartphone. Of course, it can be used for business cards or advertisements.

  The half-format uses four correction points, each located in a square of a true color image, applied only to shape position correction and simple color correction, the correction method is the same as the full format, and the correction points are black This is because the addition of a frame causes camera processing software such as edge enhancement of an environmental image approaching white to cover the area of the correction point and cause a color difference.

  In this format, there is no ID number to be displayed, and after the true color image in the true color image is converted into a true color image by the identification software in the mobile terminal such as a smartphone, the acquired true color image is converted to a server by a network. Need to be shipped to and compared, this situation means the following: 1. The search may fail. 2. Net marking must be done. 3 The reverberation speed in a timely manner is slowed down, affected by the network bandwidth, etc., the format is simple and intuitive, the influence on the original figure is small, the amount of image data is large, and there is a maximum search accuracy. There are four correction points that can be used and used for correction purposes, as well as submitting to the user that this figure is an identification diagram, and related information can be retrieved by shooting.

  Pure image format contains only true color images. The pure image format is mainly applied to the real environment. For example, the McDonald's sign, the lantern advertising billboard, etc. cannot apply both formats, but has a feature that the figure is simple and the identification is easy. If such a format is adopted, it is possible to obtain a true map by scanning, and after converting the true map into a true color image using identification software, it is sent to the server via the network and the corresponding material is searched. It ’s fine. Such a pure image format can be widely used and has a great influence on people's daily lives.

  As shown in FIGS. 4 to 6, based on the four-dimensional information, the image identification system of the present invention includes an identification diagram, a series of identification data corresponding to the identification diagram, and a server. The identification diagram is also a true color image, a two-dimensional code, a color that is superimposed on the two-dimensional code, a complete format containing an ID number and a plurality of correction points, and a true color image in which one correction point is installed in each square. Three formats, including a half-format containing a pure color image and a pure image format containing only a true color image. The server includes a receiving / shipping module, an identification data generation module, a database for storing identification data, and a search determination module.

  The identification map is acquired by the mobile terminal and sent to the server via the network. The server receives the identification map, analyzes the corresponding identification data, and compares the identification data stored in the database, and then moves the identification information. Return it to the terminal or computer. The mobile terminal shoots the identification map, that is, uses the imaging function of a smartphone, a computer, a dedicated instrument, etc. Provide data on elements, for example, analyze 2D codes, colors superimposed on 2D codes, convert true diagrams to true color images, obtain identification data or indexes, and use them for databases over networks The data is sent to the server, and the database server receives the identification data or index and compares it with the data stored in the storage device, and then returns the identification information to the mobile terminal. The object to be photographed by the mobile terminal may be one business card, one advertisement, one video, for example, a real object such as a face and a palm, a traffic sign, or a cup.

  The true color image includes a planar true color image, an uneven true color image, a real figure, or a biological figure. The color adopts RGB color standard. The identification diagram is placed in a black or colored box. The identification diagram further includes a plurality of correction points placed in the square. Mobile terminals include smartphones, IPAD (registered trademark), scanners, or notebook computers. Four-dimensional codes are issued and used by electronic documents, screens or printing. Identification data includes true color images, logos, screens, category numbers, business cards, personal business cards, and product descriptions.

  The true color image is the same as the company trademark and is displayed on electronic documents and videos in computers and smartphones, or printed on business cards, books, advertisements, or actual objects such as palm prints. The identification map is made gray scale according to outdoor operation or printing demand, and special classification processing is performed by software. In this embodiment, the color can overlap with the two-dimensional code. The color standard adopted for the color because the user uses the mode to input an electronic document by a client such as a computer on behalf of the user, and the drawing is identified by an application in a smartphone, a panel computer, a desktop computer, or a laptop computer. Is also an RGB color mode, and this RGB color mode is the color standard of the light source display system, and the three colors red (R), green (G), and blue (B) are changed, and they are overlapped with each other. Each color is acquired by being. Among them, RGB represents the colors of red, green, and blue passages, and since the overlap of light sources is adopted, the total number of RGB becomes large and the color becomes shallow. Each color is divided into 25 gradations in RGB, each of which is 0 to 255, and the three primary colors are displayed as 256 * 256 * 256 total 16777216 colors on the screen, and expressed as #FFFFFF in hexadecimal notation in programming. The

  Four-dimensional code true-color images are more widely applied in commerce. For example, when you see a favorite product, you know the manufacturer's materials by shooting, and then directly purchase and purchase an order form on your smartphone. If you suspect a counterfeit, you can capture detailed information of the genuine manufacturer by taking a 4D true color image of the 4D code, and in this way the user knows whether the purchase is fake and achieves the function of preventing forgery To do. True color images can be used for further encryption. For example, you can take a picture of your palm print and compare it with a reference map stored in a network database. This can be used as a smartphone unlock mode, or for online shopping and shopping Therefore, the encryption mode of the bank card payment is set. For this reason, if the man-machine separation or the man-card separation is used, decryption cannot be performed, and the safety of trade is improved. This method can be further used for face identification, number plate identification, traffic public facility identification, and the like. True-color images of four-dimensional code can be further encrypted by shooting with a smartphone, shooting a dynamic data chain displayed on a bank terminal immediately on a smartphone, and transferring it to a transmission bank via a smartphone network. A bank terminal may obtain a smart phone number and a data chain, and further obtain palmprint identification and fingerprint identification (provided by the bank terminal), and agrees to trade.

  However, the true color image of the four-dimensional code also has a certain locality, whereas when the image is complicated, the image noise is large, and the ambient light is dark or bright, it is easy to cause a problem of image contrast. In addition, when photographing mainly with a smartphone, a panel computer, etc., even when the lens resolution is insufficient and the camera shake is large, the identification is similarly difficult. However, these problems are caused by minimizing the various shooting situations that can be done, and providing a corresponding method, for example, shooting palmprints for encryption, and selecting and confirming whether to shoot with the left or right hand in a smartphone program. , Request that the longest handprint be oriented vertically and the correction point in the program of the smartphone be aligned with the upper end surface and the longest handprint. Similarly, there are corresponding requirements for photographing faces, products, public signs, company trademarks, etc., and the problems of image contrast and identification can be solved well.

Based on the four-dimensional information, the present invention further provides a four-dimensional code search system, including an identification diagram and a series of identification data corresponding to the identification diagram, a true color image including the identification diagram, a two-dimensional code, and a two-dimensional code. Including the color and ID number to be superimposed, the true color image, the two-dimensional code, the color to be superimposed on the two-dimensional code, and the ID number have the same or corresponding index. The server includes a receiving / shipping module, an identification data generation module, a database for storing identification data, and a search determination module. After the identification map is obtained by the mobile terminal and sent to the server by the network, the server receives the identification map, analyzes the corresponding identification data, and compares it with the identification data stored in the database. Return to the mobile terminal or computer.
The mobile terminal of the four-dimensional code search system includes a smartphone, IPAD (registered trademark), a scanner, or a notebook computer.
As shown in FIG. 7, based on the four-dimensional information, the present invention further provides a four-dimensional code search method including the following procedure:
Procedure 1. 4D code information is acquired by photographing with a mobile terminal and sent to a server via a network;

  Procedure 2. The server receives the four-dimensional code information, analyzes the corresponding identification data, and returns the identification information to the mobile terminal after being compared with the identification data stored in the server database.

  If the smartphone and database are in complete format with the ID number as the connection port, the smartphone will not have a shooting function, the shooting will be wrong, the computer will not have a low light, and only the ID number will be announced over a long distance If the situation occurs, it will cause the camera not to shoot. In this case, the user accesses the system or network database in the same manner by manually inputting the ID number through the smartphone interface or the computer interface. In other cases, the full format calculates a true color image, a two-dimensional code and / or a ternary color code by shooting, and the two-dimensional code and / or ternary color code does not follow the network. Regular information content can be displayed directly on a smartphone, and at the same time, a network database is searched by a true color image, a two-dimensional code and / or a ternary color code and an ID number.

  When the half format and the pure drawing format are adopted, a true color image is generated and can be searched only in the database, and thus cannot be identified by the deformation distortion of the drawing, and it is easy to cause a search failure in the end. For this reason, fault tolerant processing is required in a plurality of processes. The present invention provides two fault tolerant processes. First, calculation is performed by using the matrix feature of a graphic dot matrix, which will be described in detail below. Next, after the similarity calculation is performed on the data in the database of the network, and the graphic having a high similarity is sent to the interface of the smartphone, the user confirms it. Also, it is necessary to confirm the user by the smartphone user name and the smartphone user number, confirm the user's usage image, and manually enter the password or identify the anti-palmprint for encryption class operation, for example, network payment. There is.

  Since the information in the complete format is the most complete and the ID information has a third-class interactive search relationship, the complete format is preferentially used, for example, business cards, bank cards, advertisements, videos, etc. Half format can be used in a special environment, and there are four corrections to half format. When users see it, it knows that this is a 4D code format, and smartphone captures the true color image data by the system. After the calculation (conversion from graphics to digital), the corresponding ID number is retrieved by a computer, so that the user can access a specialized portal and use it for voice reading for children in this manner. For example, a user can request a pure image format, produce a business card, and capture true color image data on the business card with a user smartphone, thereby obtaining user data and accessing a user platform.

  In the process of photographing or scanning the four-dimensional code or the data box having the four-dimensional information, it is necessary to remove the influence of the photographing / scanning equipment and the environment, and the mobile device itself is responsible for the four-dimensional code or the four-dimensional information. This also has a large effect on the data frame including the above, and affects the size deformation, brightness, color saturation, and exposure level. The environment has the brightness of ambient light, and the color has a maximum effect on the data frame with the 4D code or 4D information.

  In a data box having a four-dimensional code or four-dimensional information in a publication, a part of the color pigment is severely faded by long-term irradiation with sunlight, and a gray scale identification diagram can be employed in such an environment.

A data box with 4D code or 4D information that is damaged or excessively distorted has already been partially lost and must be abandoned or searched by ID number keyed into the full format of the smartphone. The
After shooting or shooting, the noise processing is removed using software, and the processing method is as follows.
Adjacent area averaging method: Obtain a reduction in equipment noise at the expense of dummy glue.
Mid-value filtering: By adopting an intermediate value calculation or weighted arithmetic method for several courses, a reduction in equipment noise is obtained by substituting imitation glue.
Isolated noise removal: Removes independent or regular spots.
Sharpening: In other words, edge enhancement processing, the image becomes clearer, and is divided into methods such as differential sharpening and edge detection.
Color processing: automatic color tone, automatic contrast, automatic color processing (network server data is pre-processed as well to ensure consistency).

The mobile terminal or server further refines the image, and the process of image miniaturization is the process of obtaining the image framework, mainly using facial shape identification, handprint identification, company logo identification and product identification in a pure image format. To do.
The function is the same as that of the miniaturized part by adopting a method in which the mobile terminal or server further extracts the contour and makes the internal point hollow.

  The mobile terminal or server further extracts pattern features and adopts a structure analysis method and a statistical analysis method. In the structure analysis method, pattern elements are extracted by a formalized language, and the arrangement rules of the pattern elements are inferred. Mainly uses pure image format identification. The statistical analysis method is the main method used for identification diagrams, confirms the origin and end point of the graphic matrix, establishes the graphic digitization matrix, and compares it with the digitized matrix of the reference diagram to confirm the similarity And enter into evaluation. Mainly uses full format half format identification.

  The mobile terminal or server performs image identification processing on a data box having a four-dimensional code or four-dimensional information. Hereinafter, the image identification processing program will be described.

  Full-format and half-format mathematical models are established as follows. By adopting a method in which the figure dot matrix is made to correspond to the digital matrix, each figure dot matrix corresponds to the three primary color matrixes of the red digital matrix, the green digital matrix, and the blue digital matrix, respectively. For example, for primary data codes, two-dimensional codes, ternary color codes and true color images.

  Feature extraction and retrieval as a true color image from the three primary color matrix, true color image retrieval, ID number retrieval, and two-dimensional code and / or interrelation and interactive retrieval method of colors superimposed on two-dimensional code Form.

実数部分：

虚数部分：

実数部分：

虚数部分：

二次元Ｇａｂｏｒウエーブフィルター処理の試料特徴次元が更に高くなり、簡単なデータ特徴として圧縮する必要があることで、スマートフォン、コンピュータ等の演算速度を向上させるため、画像の試料特徴次元を低下させ、ここで二次元離散余弦変換方式を利用して次元を低下させる。離散余弦変換は、特にその第二種類型（写真）が常に信号処理及び画像処理に使用され、且つ信号及び画像（静止画像及び運動画像含有）に関する損壊データ圧縮に使用される。この理由は離散余弦変換が強い「エネルギー集中」特性を備える。大多数の自然信号（声音及び画像含有）のエネルギーが離散余弦変換された低周波部分に集中し、信号にマルコフ過程に接近する統計特性を備える場合には、離散余弦変換の関連解除性がＫ−Ｌ変換（Ｋａｒｈｕｎｅｎ−Ｌｏｅｖｅ変換−最優の関連解除性を具備）の性能に接近する。
例えば、静止画像コード基準ＪＰＥＧ中で、運動画像コード基準ＭＪＰＥＧ及びＭＰＥＧの各基準にそれぞれ離散余弦変換が使用される。これらの基準制に二次元の第二種類型離散余弦変換を使用し、且つ結果を量子化させた後、エントロピー符号化行う。この場合、相応的に第二種類型離散余弦変換のｎが通常８とし、該公式で各８ｘ８ブロック毎の毎行を変換した後、毎列を変換する。一つの８ｘ８の変換係数マトリックスを得る。その中で（０、０）位置の元素が直流分量とし、マトリックス内の他の元素がその位置により周波数の異なる交流分量を表示する。
二次元離散余弦変換定義が次の通りである：

Since the three digital matrices contain many patterns and the two-dimensional Gabor wave filter has a clear advantage in the pattern extraction surface, the present invention employs the two-dimensional Gabor wave filter to extract the patterns.
A two-dimensional Gabor function is formed by a synergistic product of a Gaussian function and a complex plane wave, and its two-dimensional expression is as follows.
Regression expression:

Real part:

Imaginary part:

Real part:

Imaginary part:

The sample feature dimension of the two-dimensional Gabor wave filter processing is further increased, and it is necessary to compress it as a simple data feature. This improves the calculation speed of smartphones, computers, etc. Reduce the dimension using a two-dimensional discrete cosine transform method. The discrete cosine transform is used in particular for its second type (photo), which is always used for signal processing and image processing, and for compressed data compression on signals and images (including still and motion images). This is because the discrete cosine transform has a strong “energy concentration” characteristic. If the energy of the majority of natural signals (contained voice and image) is concentrated in the low-frequency part subjected to discrete cosine transform, and the signal has a statistical characteristic that approximates a Markov process, the relativity of discrete cosine transform is K -Approach the performance of the L transform (Karhunen-Loeve transform-with the best association release).
For example, in the still image code standard JPEG, discrete cosine transform is used for each of the motion image code standard MJPEG and MPEG standards. Two-dimensional second-type discrete cosine transform is used for these reference systems, and the result is quantized and then entropy encoded. In this case, n of the second type discrete cosine transform is normally set to 8, and after converting every row of each 8 × 8 block by this formula, every column is transformed. One 8 × 8 transform coefficient matrix is obtained. Among them, the element at the (0, 0) position is a direct current quantity, and the other elements in the matrix display the alternating current quantity having a different frequency depending on the position.
The two-dimensional discrete cosine transform definition is as follows:

純粋画像フォーマット数学模型が以下のように確立される。 The graphic matrix processing method based on the discrete cosine transform employs a Z-type imaging method, and can effectively extract large coefficient information of an image. The operation process is specifically shown in the following figure.

A pure image format mathematical model is established as follows.

Specifically, when the full format or half format is used in an electronic document or printed product, the readability of the full format or half format is provided so that a calibration point is provided and the user can photograph and control significantly. Strong, pure image format-like images have no borders, no corrections, inevitably poor readability, and a few trades and algorithms increased based on full-format, half-format algorithms to improve readability These contracts are displayed as calculations in terminal tool applications such as smartphones or panel computers, and are displayed by classification and explanation in the program interface. After the calculations are identified at the same time, the number of similarities in the terminal application is determined by the sorting method. Submit a picture of To be confirmed by The. Hereinafter, an operation example will be described.
Common identification requirements for pure image formats:

  As a requirement of directionality, it is required to match the arrangement angle of the graphic product with the identification diagram, and if necessary, another visual angle identification diagram is requested and correlated with the use characteristics of the user.

  Requirement of shooting zone: The frame size of the shooting zone is the same as the frame in the smartphone application, and the installation design of the shooting target is required to compare the background with a large color and the deep and bright scene, and the background frame as much as possible And presents an identification diagram feature to attract the user to take a picture.

  Different from the full format and half format, the identification diagram of the pure image format only needs to be one type in the system, so it can be professionally designed and exhibits imaginative power, and as a result, transition between the pure image format and the half format The figure can be designed, and it is only necessary to present the user to take the picture.

  Road sign identification: A road sign is an identification chart of typical pure image format characteristics, it is easy to identify because it is fluent and simple, has no color transition zone (gradual change color), Identification is possible if the common identification requirement is satisfied.

  Face shape identification: Since face recognition uses the identification mode of the pure image format, it is also necessary to meet the common identification requirements of the pure image format, and at the same time the face image has self-specific compatibility, There is a feature that you want to search for: facial communality includes hair, eyes, eyebrows, nose, and lips, and the size and position varies depending on the individual, especially images with a current smartphone camera resolution of 10 million The difference is greater.

  Currently, face identification is relatively mature, and the present invention adopts a digital matrix corresponding to the same image dot matrix as the complete format for the high efficiency of smartphone terminal calculation, and the calculation method is the same. Only process the picture frame and positioning points. The frame of the image is the chin and both ears, both eyes are the positioning points, and the angle and positioning of the image are calculated (the image frame and the positioning points appear on the smartphone terminal). In other cases, the problem is solved by, for example, associating the identification diagram as a pair of a bun and a low head. Such a method is better than the current face identification technology. The reason is that other software calculates the right / left eye height ratio, width ratio, heel height ratio, and width ratio, and further calculates the shape of the eyebrows and ears. The most distinguishable feature of wrinkles, spots or scars is to be abandoned. The present invention obtains the maximum accuracy rate by adopting a data matrix of graphic dot matrix and employing a method of fault tolerant on the calculation surface and visual selection judgment by the user interface.

  Regarding the identification of the palm shape, as long as smartphone operation is very universal, many people love self-photographing, it is convenient to photograph their palmprint with a smartphone, and as long as you use a dedicated instrument, Things like fingerprints can be identified, and this palm print can be encrypted by the identification diagram system. Especially when a smartphone is separated from a person, a QQ number, a payment password, and a bank card are stolen. Palmprint identification can reach the level of palm shape identification, finger shape identification, and the like.

In order to improve the feasibility of encryption operation, the algorithm increased for palmprint identification increases the algorithm of the digital contour feature and the digital skeleton feature based on the matrix feature proposal method. The calculation method of palm shape identification is the same as face shape identification. An image frame and positioning points for palm surface identification can be taken after the left and right hands are confirmed, as shown in the right figure (the image frame and positioning points are displayed on the smartphone terminal).
Identification of company signs is the same as identification of road signs.

  Product identification is similar to face identification, and has the most typical features and obvious features in photography, and is required to solve the problem by a method that constructs related identification diagrams by multiple identification diagrams. Shown below. Use the full format or half format as much as possible for products that the company itself wants to popularize in advertising, in particular the full format is searched with 100% net marking and includes relevant company materials.

  A four-dimensional code is an organic combination of data and an image. An image is acquired by data, data is also acquired by image, converted to an image by data, and feature data is detected by the image. The four-dimensional code defines only one interface standard, and forms a public platform with one big data feature, mutual search and identification of multiple figures formed by this standard, and deploys operations to each industry.

  Since smartphones are equipped with a camera for smartphones that can be photographed at any time, an identification map is displayed. Therefore, a smartphone portal based on the identification map has been established. For example, company identification (corresponding to the second trademark of the company), personal identification (corresponding to personal FACE or ID), product or work identification (for example, product external identification or credit card with personal portrait) ).

  Protective action on product appearance patents will preferentially accept appearance patent products, and after products with the same appearance are photographed and identified, they will preferentially access the corporate portal and violate Chinese patent law requirements due to differences in details Issue fake inferior goods for things.

  It solves the security problem of payment of a smartphone or computer network, for example, when a smartphone is lost, it ensures that another person cannot steal the money by palmprint identification.

  The four-dimensional code may be the second trademark of the company company, the company trademark needs to be approved by the National Trademark Office, is difficult to approve due to the complexity of the flow, and only black and white documents. The four-dimensional code is managed by a private company, is operated under the premise that it does not violate the trademark law, is classified by digital ID, and can be directed to the portal directly by computer and smartphone APP. It has the following advantages: true colors, complex designs, direct portal orientation, easy to remember, no input required, and relevance to product codes. The four-dimensional code is developed and controlled by a private company, establishes formal standards, and provides services to society. The registered trademark is filed in the state by the owning company, the four-dimensional code is filed by the other company in the four-dimensional code company, the trademark is solemn and seriously expressed, not wrong, and the four-dimensional code is agile Is understood to be close. Trademark features are used for distinction only, and four-dimensional codes are used for rapid identification and quick access.

  Four-dimensional code is rich in color, designers are not restricted freely and can be designed with all four-dimensional code, and the four-dimensional code is compatible with modern society, and also the current network, computer, Fits smart phones and makes it convenient to use business cards, bus cards, and bank cards. In addition, you can shoot four-dimensional codes on TV and advertisements. For example, you can shoot a McDonald's sign using a smartphone at a distance of 100 meters and add GPS to access this fast food restaurant portal. I know that the hamburger is just sold out and I don't have to go buy it.

  Operation of face: A) Operation to Meisei (applicable to all people): After establishing the identification map about Meisei in the standard formula, if you take a picture of a city or movie TV, you can access this Meisei portal immediately. May interact with you. B) Operation for nomination of crime offenses: After the public security department has established an identification chart, anyone can arrest a crime offender as well.

  Product identification diagram: If there is no packaging or code in one empty bottle, you can know what liquor bottle by taking a picture if you establish a product identification diagram. The product identification diagram is an identification diagram generated by photographing a real object. Furthermore, no matter how small the product is in one grain, it can be identified if the appearance of such rice can be distinguished from other rice.

  Protective action on product appearance patent: Accepting appearance patent product preferentially, after the product with the same appearance is photographed and identified, prioritize accessing the corporate portal and violate Chinese patent law requirements due to different details If so, it can issue a fake inferior product. This identification diagram is also used as a counterfeit instrument.

  Multimedia operations: After creating identification diagrams for products such as TV series, movies, music, and book software games, the audience sees the identification diagrams and knows what to do, and what is interesting If you shoot this with any medium, you can watch it and listen to it. If you are satisfied, you can buy it immediately.

  Logistics operation: If you establish a logistics identification chart for products, you will see the contents of the packaging for warehouse storage and logistics as a whole, and if you take a picture, you will have access to logistics control and unified big data No contractual commitment is required. The whole logistics is more unified and rapid, and can interact well with users.

  Establishment of national identification map identification system: After the identification map identification system was established with public facilities and signs, the sign information can be acquired when the smartphone for public use captures the road sign, and when the bus number and road number are photographed, the bus You can access the portal and know when the next bus will arrive and how many people will wait. After the passenger cars are made intelligent, they are automatically presented with automatic shooting signs.

  Educational impact: For example, in the case of early childhood education, a book composed of identification diagrams can be photographed with a smartphone, and the corresponding reading and interpretation can be obtained, and the corresponding presentation and answer can be obtained by photographing geometric figures and chemical equations. it can. These figures cannot be solved by conventional methods and, as another example, can be applied correspondingly to a theme park or temple.

  Operation on the encryption surface: Since the identification diagram itself is figure identification, the identification diagram itself is encrypted, and for example, an identification diagram is created for a palm print, a fingerprint, and a bill, and digital numbers and APP encryption are added, Multi-layer passwords, forming a sufficiently large number of encryption layers, improving the encryption capability, for example, adding an identification diagram to the bank card, making it easy to separate the bank card code from the bank card, on the user's smartphone It turns out that the picture is taken, and this identification chart itself is another encrypted password for the bank card. For example, when paying a treasure, in addition to entering a password, the user's own smartphone must take a picture of his / her palm print, thus not worrying about the problem of theft of funds due to the loss of the smartphone.

  Document, software, video and audio sharing points: Using the cloud storage characteristics of identification diagrams, complete the transfer and sharing of documents, software, video and audio between smartphones and computers. Can be registered to the ID, generate a corresponding identification diagram, shoot with the smartphone, and then transfer to the smartphone. Or, in the latter half of the year, a net group of identification diagrams was established, which produced the same functions as QQ, and the interaction between the smartphone and the computer was stronger, chatting with one drawing, sending and receiving documents, and video and audio with software Can be shipped and includes access to the portal.

  Interactive identification between smartphone and entrance guard system: If the code value of the identification map is input to the entrance guard system, the identification map displayed on the smartphone can be authorized in this way, and if both have a camera and a screen, interactive Identification can be realized.

  Code wireless verification between electronic products is solved, for example, Bluetooth and WIFI identification codes can be entered, and code verification can be performed by taking a picture toward a Bluetooth speaker, for example. it can.

  These parts are realized with a two-dimensional code, and the identification diagram is not only more aesthetically embracing the data, but also as a quick search criterion for graphicization, the identification diagram can be searched rapidly according to big data , You can quickly connect the website and access the portal.

  With the above technical scheme, generation of a four-dimensional code is completed, and a smart phone, a panel computer, a personal computer or the like is used to search interactively by relying on each dimension information of the four-dimensional code.

  The above description uses only specific embodiments to illustrate the present invention. Appropriately pointed out to those skilled in the art that further modifications and changes can be made without departing from the concept of the present invention, and these modifications and changes should fall within the protection scope of the present invention.

Claims (15)

Full format with true color image, 2D code, color superimposed on 2D code, ID number and multiple correction points, and half format with true color image with one correction point in each square An identification diagram including at least a pure image format containing only true color images and a series of identification data corresponding to the identification diagram;
A server including a receiving / shipping module, an identification data generation module, a database for storing identification data, and a search determination module;
The identification diagram is obtained by the mobile terminal and sent to the server by the network, the server receives the identification diagram, analyzes the corresponding identification data, and compares it with the identification data stored in the database. An image identification system for returning information to a mobile terminal or a computer.   In the complete format, the two-dimensional code is arranged above the true color image, the ID number is arranged below the true color image, adopting eight color correction points, and the color correction points are respectively The image identification system according to claim 1, wherein four are arranged in the upper left, two in the upper right, one in the lower left, and one in the lower right.   The image identification system according to claim 1, wherein a color standard of the identification diagram employs an RGB color standard.     The image identification system according to claim 1, wherein the true color image includes a planar true color image, an uneven true color image, a real figure, or a biological figure.   The image identification system according to claim 1, wherein the identification data includes a true color image, a logo, an ID number, a category number, a company business card, a personal business card, and a product content description.   The image identification system according to claim 1, wherein the identification diagram is installed in a black or colored enclosure.   The image identification system according to claim 1, wherein the identification diagram is issued and used by an electronic document, a screen, or printing.   The image identification system according to claim 1, wherein the mobile terminal includes a smartphone, an IPAD (registered trademark), a scanner, or a notebook computer.   The identification data generation module is used to complete the generation and overlap of true color image data, two-dimensional code data, color and ID number, a true color image processing sub-module, a two-dimensional code data processing sub-module, a color The image identification system according to claim 1, further comprising a processing submodule and an ID number processing submodule.   10. The image identification system according to claim 9, wherein the true color image processing sub-module processes a true color image, and the processing includes geometric transformation, noise removal, miniaturization, contour extraction, and pattern feature extraction. .   The image according to claim 1, wherein the search determination module includes a true color image data conversion submodule, a two-dimensional code data conversion submodule, a color conversion submodule, an ID number conversion submodule, and a contrast submodule. Identification system. Procedure 1 for obtaining identification diagram information by photographing with a mobile terminal and sending this information to a server via a network;
Step 2 in which the server receives the identification map information, analyzes the corresponding identification data, and searches and compares the identification data stored in the server database, and then returns the identification information to the mobile terminal.
An image identification method comprising:   An identification diagram and a series of identification data corresponding to the identification diagram, wherein the identification diagram includes a true color image, a two-dimensional code, a color superimposed on the two-dimensional code and an ID number, and the true color image, the two-dimensional A four-dimensional code characterized in that the code, the color superimposed on the two-dimensional code, and the ID number have the same or corresponding indexes. A true color image, each having a similar or corresponding index, a two-dimensional code, an identification diagram including colors and ID numbers superimposed on the two-dimensional code, and a series of identification data corresponding to the identification diagram;
A receiving / shipping module, an identification data generation module, a database for storing identification data, and a server including a search determination module;
After the identification map is obtained by the mobile terminal and sent to the server by the network, the server receives the identification map, analyzes the corresponding identification data, and compares it with the identification data stored in the database. A four-dimensional code retrieval system characterized in that a message is returned to a mobile terminal or a computer. Procedure 1 for photographing with a mobile terminal to obtain four-dimensional code information and sending it to a server via a network;
Step 2 in which the server receives the four-dimensional code information, analyzes the corresponding identification data, and searches and compares the identification data stored in the database of the server, and then returns the identification information to the mobile terminal.
Including four-dimensional code search method.

Images