CN108805246B - Intelligent indication two-dimensional code and method for monitoring food state by using same - Google Patents

Abstract

The invention discloses an intelligent indicating two-dimensional code and a method for monitoring food state by using the same, wherein the two-dimensional code comprises a position detection graph, a data information graph, an indicating information graph, a format information graph and a position detection graph divider; the position detection graph can be used for positioning the position of the two-dimensional code; recording product information by data information graph; one piece of the indicating information graph is coated with an indicator, and the other piece of the indicating information graph is printed with a standard color corresponding to the indicator; the format information graph indicates format information and version information of the data; the position detection pattern splitter is used to separate the position detection pattern from the format information and data information patterns. The terminal scans the two-dimensional code to obtain the information of each graph, the indicator information is input into the model to obtain the color grade and is sent to the server, and the server determines the state of the product and sends the state of the product back to the terminal. The product quality state is reflected according to the color change condition of the indicator of the two-dimensional code, and the product state in transportation and storage is monitored in real time by combining the intelligent terminal and the server.

Description

A kind of intelligent indicating two-dimensional code and method for monitoring food state by using the two-dimensional code

technical field

The invention belongs to the technical field of two-dimensional code preparation, and in particular relates to an intelligent indicating two-dimensional code and a method for monitoring food quality by using the intelligent indicating two-dimensional code.

Background technique

With the acceleration of urbanization in my country to drive urban distribution, the explosive growth of the online retail market, the outsourcing of manufacturing logistics links and the promotion of the cold chain logistics market of the fresh food industry, my country's logistics industry has ushered in rapid growth in recent years. According to China Industry Research According to the China logistics survey released by the Internet, the total social logistics in the country in 2016 was 229.7 trillion yuan, and the market size of the logistics industry is expected to be close to 360 trillion yuan by 2020. Fresh food, fruit and vegetable food and other products are prone to spoilage and corruption during the logistics process. It is particularly important to record the source information of such products and the quality information after transportation and storage. The technology that has been popularized to record product information is barcode technology.

One-dimensional barcode (One-Dimensional Bar Code) is the most frequently used in recording product information. The application of one-dimensional barcode can improve the speed of information entry and reduce the error rate, but one-dimensional barcode also has some shortcomings: the data capacity is only The barcodes that can contain letters and numbers are relatively large in size and have low space utilization, and the barcodes cannot be read after being damaged.

Two-dimensional code (QR Code) is a black-and-white graphic recording data symbol information that uses a specific geometric graphic to distribute in the two-dimensional direction of the plane according to a certain rule. One-dimensional codes have many features. As an information storage, transmission and identification technology, two-dimensional code is not only used in the automatic management of industrial production lines, the classification of goods in logistics, the query of inventory, the management of postal logistics and parcels, etc., but also in customs, taxation and other departments. It can be applied to the management of similar reports and bills, as well as the management of various certificates, but the QR code records a small amount of information in the amount of information. When the amount of information is too large, it cannot be recognized and read, and the compilation efficiency is low.

The barcodes in the prior art that can record product information, whether it is a one-dimensional code or a two-dimensional code, can only record the pre-recorded information, but cannot reflect the specific quality status information of the product after transportation and storage in real time.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome the defects existing in the prior art: the currently applied one-dimensional code and two-dimensional code can only record pre-written information, and cannot reflect the specific quality state of the product in real time after the complex transportation and storage process In view of the shortcomings of information, the present invention provides an indicative two-dimensional code and an identification method that can monitor changes in product status during transportation and storage, so that the time-temperature of food during transportation and storage can be obtained by scanning the two-dimensional code through a terminal device. , humidity, pH, oxygen, information on microbial quality indicators.

The invention provides an intelligent indicating two-dimensional code, wherein the structure of the intelligent indicating two-dimensional code is that a data layer is wrapped in the middle of two protective layers. The protective layer is a transparent plastic film with a thickness of 20-25 microns, and the material can be nylon plastic film, polyester plastic film or o-phenylphenol plastic film. The data layer material is non-porous and breathable material such as white paper or polyester non-woven fabric. The intelligent indicating QR code is located on the data layer, including position detection graphics, data information graphics, instruction information graphics, format information graphics and position detection graphics separators. The upper protective layer and the lower protective layer corresponding to the position of the indication information graphic are evenly distributed with air-permeable micropores with a diameter of 0.3-1.0 mm.

The location detection patterns include a first location detection pattern, a second location detection pattern, and a third location detection pattern, which are distributed in the upper left corner, the upper right corner and the lower left corner of the two-dimensional code, respectively.

The data information graphic is used to record binary data, and the binary data records the origin information, production information and indication type information of the product;

The indication information graphic occupies 2 data blocks, and the 2 data blocks are adjacent and located in the middle of the smart indication QR code. The first data block is coated with an indicator, and the indicator can be time-temperature indicator, humidity indicator, pH indicator One or more of the value indicator, oxygen indicator or microbial indicator, each indicator can display different colors according to the corresponding environmental conditions; the second data block is printed with corresponding to the adjacent first data block The standard color of the indicator, the standard color is the combination of the color of the indicator in different stages of change.

Above-mentioned time-temperature indicator is octadecyl urethane, butoxycarbonyl urethane or starch/amylase;

Above-mentioned humidity indicator is copper chloride or cobalt thiocyanate;

The above-mentioned pH indicator is anthocyanin, methyl orange or neutral red;

Above-mentioned oxygen indicator is o-phenanthroline-Fe(II), N-o-anilino acid or sodium diphenylamine sulfonate;

The above microbial indicators are EZTest@ethylene oxide or EZTest@hydrogen peroxide.

The format information graphic is used to indicate the format information and version size information of the data during encoding. The format information of the two-dimensional code stores the error correction level of the two-dimensional code. The two-dimensional code in the present invention adopts the M-level error correction code. Character.

Among them, the error correction code word is calculated by the data code word using the Reed-Solomon error correction algorithm based on the selected error correction level.

The position detection graphic separator is used to separate the position detection graphic from the format information, or separate the position detection graphic from the data information graphic.

The present invention provides a method for identifying an indicative two-dimensional code, the method comprising:

Step 1, obtain the image of the two-dimensional code;

Step 2, identify 3 position detection patterns in the two-dimensional code;

Step 3, read the format information and version information in the two-dimensional code format information graphics;

Step 4, decode the data code word from the data block;

Step 5, obtaining the graphic position of the indicator indicating the information graphic and the standard color;

Step 6, correcting the graphic information of the indicator;

Step 7, match the indicator color grade;

Step 8, sending the level information of the indicator to the server;

Step 9, the product status information fed back by the server to the terminal.

Data code word in described step 4, it records the kind information of indicator, product origin information, product production information;

In the step 5, the position of the indicator indicating the information graphic and the graphic position of the standard color is obtained, specifically: in order to improve the image processing efficiency, the image is first converted into a gray value image with a gray value of 32. Use the median filtered 5×5 template to remove noise from the gray value image, which is used to enhance the image effect and facilitate post-processing; then use the Sober operator to detect the edge of the image to obtain the edge of the indication information image, and then use the contour extraction algorithm to detect The outline of the indicator and standard color indicating the information graphic is obtained; finally, the graphic position of the indicator and the standard color is segmented by the adaptive threshold selection method.

In the step 6, the indicator graphic information is corrected, the R, G, and B values of each pixel in the indicator and the standard color graphic are extracted respectively, and the color of each area is corrected by the polynomial regression method with online correction.

Matching the color grade of the indicator in the step 7, specifically, bringing the corrected R, G, B values of the indicator into the pre-established artificial neural network model to identify the color grade of the indicator;

In the artificial neural network model established above, the number of neurons in the input layer is 3, which is the R, G, and B values corrected by the indicator, the hidden layer is 8 neurons, and the output layer is 4 neurons, which is 4 neurons. The transfer function of the hidden layer is tansig(), the transfer function of the output layer is logsig(), and the training function is trainlm().

The product status information fed back by the server to the terminal in the step 9,

Optionally, when the indicator is a time temperature indicator, the product status information is the storage time of the product;

Optionally, when the indicator is a humidity indicator, the product status information is the moisture content of the product storage environment;

Optionally, when the indicator is a pH indicator, the product status information is the pH value of the product storage environment;

Optionally, when the indicator is an oxygen indicator, the product status information is the oxygen content of the product storage environment;

Optionally, when the indicator is a microbial indicator, the product status information is whether the product is contaminated with bacteria in storage.

Beneficial effects of the present invention:

The invention relates to an intelligent indicative two-dimensional code, which is printed on the two-dimensional code by using a color-changing indicator, the indicator can change in color after the change of the external environment, and the quality state of the product can be reflected according to the changed color, so as to realize monitoring The function of product quality status in the process of transportation and storage overcomes the shortcomings of traditional barcodes that can only record fixed information, enriches the functions of two-dimensional codes, and improves the intelligence of two-dimensional codes.

Description of drawings

Fig. 1 is the structural decomposition schematic diagram of the intelligent indication two-dimensional code of the present invention;

Fig. 2 is one of the schematic diagrams of the data layer of the intelligent indication two-dimensional code of the present invention;

Fig. 3 is the second schematic diagram of the data layer of the intelligent indication two-dimensional code of the present invention;

4 is a schematic diagram 2 of an indication image of an intelligent indication two-dimensional code of the present invention;

In the figure: 101-first protective layer, 102-data layer, 103-second protective layer, 104-breathable micropores, 201-position detection pattern, 201a-first position detection pattern, 201b-second position detection pattern, 201c-third position detection graphic, 202-data information graphic, 203-format information graphic, 204-position detection graphic separator, 205-indication information graphic, 205a-first data block, 205b-second data block.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

Example 1:

The structure of an intelligent indication two-dimensional code proposed by the present invention is shown in FIG.

The first protective layer 101 and the second protective layer 103 are transparent plastic films with a thickness of 20-25 microns, and the materials may be nylon plastic films, polyester plastic films or o-phenylphenol plastic films. The protective layer has a wide operating temperature range, good dimensional stability, corrosion resistance, and non-toxicity, and protects the data layer 102 from physical wear or scratches. QR code image.

The data layer 102 is a non-porous and breathable material such as white paper or polyester non-woven fabric, and a two-dimensional code is printed on the data layer 102. The two-dimensional code includes a position detection graphic 201, a data information graphic 202, an instruction information graphic 205, and a format. The information graphic 203 and the position detection graphic separator 204 indicate that an indicator that can react with the external environment is printed on the information graphic 205, and the upper protective layer and the lower protective layer corresponding to the position of the information graphic 205 are evenly distributed with a pore size of 0.3- The breathable micropores 104 of 1 mm are convenient for the indicator to contact with the outside world.

Example 2:

In order to illustrate the structure of the indicative two-dimensional code in more detail, the following embodiments are described by taking a two-dimensional code with version 1 as an example.

A schematic structural diagram of a two-dimensional code data layer provided by an embodiment of the present invention is shown in FIGS. 2 and 3 . The two-dimensional code is a two-dimensional code with 21×21 symbols, including a position detection graphic 201 , a data information graphic 202 , an instruction information graphic 205 , a format information graphic 203 and a position detection graphic separator 204 .

(1) The location detection pattern 201 includes a first location detection pattern 201a, a second location detection pattern 201b and a third location detection pattern 201c, which are respectively distributed in the upper left corner, upper right corner and lower left corner of the two-dimensional code.

(2) The data information graphic 202 occupies 24 data blocks. The data information graphic 202 is used to carry binary data, and the binary data records the origin information, production information and indication type information of the product;

(3) The indication information graphic 205 occupies 2 data blocks, and the 2 data blocks are adjacent and located in the middle of the two-dimensional code, wherein the first data block 205a is coated with an indicator, and the indicator can be a time-temperature indicator, a humidity indicator One of the indicator, pH indicator, oxygen indicator or microbial indicator, each indicator can display different colors according to the corresponding environmental conditions; the second data block 205b is printed with instructions corresponding to the first data block 205a The standard color of the indicator, the standard color is the combination of the color of the indicator in different stages of change, as shown in Figure 4, 205b1, 205b2, 205b3 and 205b4, which are the intermediate colors within the discoloration range of the indicator. The function of the standard color is to eliminate the Light-induced color errors and matching to indicator change levels.

The time-temperature indicator can be a mixture of starch and amylase that changes color from white to blue over time;

The humidity indicator can be a mixture impregnated with ferric or ferrous salts, the color of which changes from yellow/amber to almost colorless with increasing humidity;

The pH indicator can be purple potato anthocyanin, and its color can change from red to blue as the pH changes from acidic to alkaline;

The oxygen indicator can be, o-phenanthroline-Fe(II), which itself is light blue and turns red after contact with oxygen;

The microbial indicator can be EZTest@ethylene oxide, the color is red when sterile, and the color is yellow when microorganisms are present;

(4) The format information graphic 203 is shown in the shaded part of the oblique line in FIG. 2, and is used to indicate the format information and version size information of the data during encoding. The format information of the two-dimensional code stores the error correction level of the two-dimensional code. The two-dimensional code in the embodiment adopts the error correction code word of M level;

(5) The position detection graphic separator 204 is used to separate the position detection graphic 201 from the format information graphic 203 , or separate the position detection graphic 201 from the data information graphic 202 .

Example 3: Identification method

The identification method of the QR code includes:

Step 1, obtain the image of the two-dimensional code;

The specific method for acquiring the image containing the two-dimensional code by the smart terminal includes placing the image containing the two-dimensional code in the identification frame, or selecting the image containing the two-dimensional code from the album, or long pressing the image containing the two-dimensional code.

Step 2, identify 3 position detection patterns in the two-dimensional code;

The intelligent terminal scans the image progressively from top to bottom.

The position detection pattern is a black and white "return" pattern, and the ratio of black and white code elements during scanning is 1:1:1:3:1:1. In the process of scanning the two-dimensional code, first search for two-dimensional The positioning pattern of the code, when all the lines in the image are scanned, n candidate position detection patterns are obtained.

The intelligent terminal connects the centers of each of the three candidate position detection patterns to obtain the corresponding triangle; determines the triangle with the highest similarity to the isosceles right triangle and the longest side; from the three candidate positions corresponding to the triangle In the detection pattern, a first position detection pattern in the upper left corner, a second position detection pattern in the upper right corner and a third position detection pattern in the lower left corner are determined. The position of the two-dimensional code in the image can be located by using three position detection patterns.

Step 3, read the format information and version information in the two-dimensional code format information graphics;

When the square module array is a square module array of 21×21 symbols, the corresponding coding format information determines the data information pattern.

Obtain the error correction level corresponding to the error correction codeword from the format information, and the error correction level includes L error correction level, M error correction level, Q error correction level and H error correction level;

Step 4, decode the data code word from the data block of the data information graphic to obtain the type information of the indicator, the product origin information, and the product production information;

Optionally, when the square module array corresponding to the two-dimensional code has a defective part and interference information exists in the square module array, the error correction codeword is used to correct the data error.

It should be noted that the error correction codeword of the present invention can correct two types of errors, refusal to read errors and substitution errors. The refusal to read error refers to a piece of data that is not scanned or cannot be decoded when the location of the erroneous data is known; the substitution error is the data that is erroneously decoded when the location of the erroneous data is unknown. For example, when a defect turns a dark module into a light module, or a light module into a dark module, i.e. incorrectly decodes data as apparently valid, but is another different data. Data substitution errors require two error correction codewords to correct.

Step 5, obtaining the graphic position of the indicator indicating the information graphic and the standard color;

In order to obtain the position of the indicator and standard color indicating the information graphic in the two-dimensional code, the image is first converted into a gray value image with a gray value of 32. Use the median filtered 5×5 template to remove noise from the gray value image, which is used to enhance the image effect and facilitate post-processing; then use the Sober operator to detect the edge of the image to obtain the edge of the indication information image, and then use the contour extraction algorithm to detect The outline of the indicator and standard color indicating the information graphic is obtained; finally, the graphic position of the indicator and the standard color is segmented by the adaptive threshold selection method.

Step 6, Correction of indicator graphic information

The R, G, and B values of each pixel are extracted from the indicator and standard color graphs segmented in step 5, respectively, and the color of each area is corrected by the polynomial regression method with online correction. The present invention adopts the polynomial regression method to make the storage space less, the calculation speed fast, and easy to implement. The following describes the color correction process based on multiple linear regression:

There are 4 color blocks on the standard color image, and the color R, G, and B values of the ith color block are _Roi , _Goi , _Boi in the standard space, where i=1, 2, 3, 4, in natural If the color R, G, and B values of the standard color image are collected under the lighting environment, then:

The matrix form of formula (1) is:

X=A ^T *V (2)

Where: X is the color scale standard R, G, B value matrix with dimension 3×i;

A is a conversion coefficient matrix of dimension j × 3;

V is a polynomial regression matrix of dimension j×i;

Among them, v _ji (j=1,...,J) consists of polynomials, there are many different polynomial forms, for example, v _ji =[R,G,B,1], v _ji =[R,G,B, RG, RB, GB, 1], v _ji = [R, G, B, RGB, 1], etc. The form of v _ji can be combined into different forms as required.

The matrix A is optimized by the least square method, and A is the required model parameter.

A=(V×V ^T ) ^-1 (V×X ^T ) (3)

Bringing A into formula (2), the R, G, and B values of each pixel of the corrected image can be calculated to realize online color correction.

X _out = A ^T ×V _in (4)

Among them, X _out is the R, G, B matrix of the corrected image, and the dimension is 3×M; V _in is the matrix formed by the terms of the polynomial corresponding to the R, G, and B values of all pixels of the original image, and the dimension is is j×M; M is the total number of pixels of the original image.

Step 7: Match Indicator Color Grades

Bring the corrected R, G, B values of the indicator into the pre-established artificial neural network model to identify the color grade of the indicator.

This embodiment uses artificial neural network modeling to match the color grade of the indicator, which has a three-layer structure, namely the input layer, the hidden layer and the output layer, each layer has multiple neurons, and the neurons in each layer pass the excitation function. For correlation, the excitation functions used include step function hardlim(x), linear transfer function purelin(x), logarithmic sigmoid function logsig(x) and tangent sigmoid function tansig(x).

To establish a color grade recognition model of artificial neural network, the process is as follows:

(1) The number of neurons in the input layer is 3, corresponding to the R, G, and B values corrected by the indicator, the hidden layer is 8 neurons, and the output layer is 4 neurons, corresponding to 4 color levels;

(2) The input matrix is normalized so that the data is distributed in the [-1,1] interval;

(3) The transfer function of the hidden layer is tansig(), the transfer function of the output layer is logsig(), and the training function is trainlm(); the maximum number of training times is 1000, and the training accuracy is 0.01.

In the process of identifying the two-dimensional code, the corrected indicator R, G, and B values are brought into the artificial neural network model, and the color grade of the indicator can be identified.

Step 8, send the color grade information of the indicator to the server, and the server determines the product status information;

The terminal is used to send the level information of the indicator to the server, and the server stores the preset correspondence between the different level information of each indicator and the product status information. information obtained by the algorithm.

Step 9, the server feeds back product status information to the terminal.

Optionally, when the indicator is a time-temperature indicator, the product status information is the storage time of the product;

Optionally, when the indicator is a humidity indicator, the product status information is the moisture content of the product storage environment;

Optionally, when the indicator is a pH indicator, the product status information is the pH value of the product storage environment;

Optionally, when the indicator is an oxygen indicator, the product status information is whether the product storage environment contains oxygen;

Optionally, when the indicator is a microbial indicator, the product status information is whether the product is contaminated with bacteria in storage;

Further, when there are multiple indicators in the instruction information image of the two-dimensional code, the product status information displays the product status information corresponding to the multiple indicators;

Example 4:

When the smart indicator QR code is used, it is installed inside the packaging box. There is a transparent window on the top side of the packaging box with the same size as the smart indicator QR code. The smart indicator QR code is pasted on the inner surface of the transparent window of the packaging box. The data layer 102 Facing outward, use the terminal to scan the smart indicator QR code through the transparent window, and then obtain the status information of the product.

The series of detailed descriptions listed above are only specific descriptions for the feasible embodiments of the present invention, and they are not used to limit the protection scope of the present invention. Changes should all be included within the protection scope of the present invention.

Claims (3)

1. a method utilizing intelligent indication two-dimensional code to monitor food quality, is characterized in that, comprises the steps: Step 1, obtain a two-dimensional code image; Step 2, identify 3 position detection patterns in the two-dimensional code; Step 3, read the format information and version information in the two-dimensional code format information graphic; Step 4, decodes the data code word from the data block of the data information graphic; Step 5, obtaining the graphic position of the indicator indicating the information graphic and the standard color; Step 6, correcting the graphic information of the indicator; Step 7, match the indicator color grade; Step 8, sending the level information of the indicator to the server, and the server determines the product status information; Step 9, the product status information fed back by the server to the terminal; The specific method of the step 5 includes: First convert the image into a gray value image with a gray value of 32; then use the median filtered 5×5 template to remove noise from the gray value image; then use the Sober operator to detect the edge of the image to obtain the indication information image. edge, and then detect the outline of the indicator and the standard color indicating the information graphic with the help of the contour extraction algorithm; finally use the adaptive threshold selection method to segment out the graphic position of the indicator graphic and the standard color; The specific method of step 6 includes: Extract the R, G, and B values of each pixel from the indicator graph and standard color graph segmented in step 5, and use the polynomial regression method to correct the color of each area, specifically: Step 6.1, there are 4 color blocks on the standard color image. Let the color R, G, and B values of the ith color block be _Roi , _Goi , _Boi in the standard space, and the standard collected in the natural lighting environment The i-th color block color R, G, and B on the color image are Ri, Gi, Bi, where i=1, 2, 3, 4, then there are: _Roi = a ₁₁ v _1i +a ₁₂ v _2i +…+a _1j v _ji G _oi =a ₂₁ v _1i +a ₂₂ v _2i +…+a _2j v _{ji Boi} = a ₃₁ v _1i +a ₃₂ v _2i +…+a _3j v _ji Wherein, v _ji =[R, G, B, 1], j = 1, ..., J; Step 6.2, write the expression of step 6.1 in matrix form as: X=A ^T *V Among them: X is the color scale standard R, G, B value matrix with dimension 3×i; A is the conversion coefficient matrix with dimension j×3;

V is a polynomial regression matrix of dimension j×i;

Step 6.3, the matrix A is obtained by using the least squares method to optimize, and A is the required model parameter; A=(V×V ^T ) ^-1 (V×X ^T ) Step 6.4: Bring A into the expression of step 6.2, and then the R, G, and B values of each pixel of the corrected image can be calculated to realize online color correction: X _out = A ^T ×V _in Among them, X _out is the R, G, B matrix of the corrected image, and the dimension is 3×M; V _in is the matrix formed by the terms of the polynomial corresponding to the R, G, and B values of all pixels of the original image, and the dimension is is j×M; M is the total number of pixels of the original image. 2. the method for food quality monitoring according to claim 1, is characterized in that, the concrete method of described step 2 comprises: The position detection pattern is a black and white loop pattern, and the ratio of black and white symbols is 1:1:1:3:1:1 during scanning. In the process of scanning the two-dimensional code, the positioning of the two-dimensional code is first searched based on this ratio. pattern, when all lines in the image are scanned, n candidate position detection patterns are obtained; Connect the centers of each of the three candidate position detection patterns to obtain the corresponding triangle; determine the triangle with the highest similarity to the isosceles right triangle and the longest side; detect the pattern from the three candidate positions corresponding to the triangle , the first position detection pattern in the upper left corner, the second position detection pattern in the upper right corner, and the third position detection pattern in the lower left corner are determined; that is, the three position detection patterns are used to locate the position of the two-dimensional code in the image. 3. the method for food quality monitoring according to claim 1, is characterized in that, the concrete method of described step 7 comprises: Bring the corrected indicator R, G, B values into the artificial neural network model to identify the color grade of the indicator, and the establishment process of the artificial neural network model is as follows: Step 7.1, the number of neurons in the input layer is 3, corresponding to the R, G, and B values corrected by the indicator, the hidden layer is 8 neurons, and the output layer is 4 neurons, corresponding to 4 color levels; Step 7.2, the input matrix is normalized, so that the data is distributed in the [-1, 1] interval; Step 7.3, set the transfer function of the hidden layer to tansig(), the transfer function of the output layer to be logsig(), and the training function to be trainlm(); the maximum number of training times is 1000, and the training accuracy is 0.01; Step 7.4, after the training is stopped, a model that can identify the color level of the indicator is obtained.

Images