WO2016206446A1

WO2016206446A1 - Character encoding method and character decoding method having error correction function and product thereof

Info

Publication number: WO2016206446A1
Application number: PCT/CN2016/079114
Authority: WO
Inventors: 吴林; 崔征
Original assignee: 信码互通（北京）科技有限公司
Priority date: 2015-06-26
Filing date: 2016-04-12
Publication date: 2016-12-29
Also published as: CN104899612A

Abstract

A character encoding method and character decoding method having an error correction function and a product using the character encoding method. The character encoding having the error correction function consists of a data character and an error correction character corresponding thereto. Using the character encoding to mark a product can effectively address the problem in the prior art in which encoded data only comprising a data character is easily misread. By determining whether the identification of the character encoding has an error via an error correction algorithm, the present invention can ensure that only the data character having been successfully corrected can be stored in the system, and the data character having been unsuccessfully corrected can be all eliminated, thereby ensuring no misread data in the data system.

Description

Character encoding method with error correction function, character decoding method and product thereof

Technical field

The present invention relates to a character encoding method and a character decoding method having an error correcting function, and to a product using the above character encoding method.

Background technique

With the development of computer technology, the processing speed of computers has been greatly improved, and computer vision recognition technology (OCR) has also been rapidly developed. These technologies are widely used in a variety of automated production lines to significantly increase production efficiency and reduce labor costs.

At present, many companies set specific character codes on products for internal management and stock management. To ensure the validity of these character encodings, companies often need to use OCR technology to read these character encodings before they leave the factory.

Although the current OCR technology is relatively mature, the accuracy rate can reach more than 90%. The printing effect is better, and the recognition rate can reach more than 98%. However, OCR technology itself has the possibility of misreading, and it is impossible to discriminate whether it is misread after OCR recognition. For example, when the number 8 in the character encoding is read as the letter B, the recognition system cannot distinguish whether or not the reading is wrong. For the existing character encoding containing only data characters, due to the misunderstanding of OCR technology, the above application system is flawed, and it is impossible to use such character encoding as a mark for the production of some systems with high precision requirements.

Summary of the invention

The primary technical problem to be solved by the present invention is to provide a character encoding method with an error correction function.

Another technical problem to be solved by the present invention is to provide a character decoding method with an error correction function.

Another technical problem to be solved by the present invention is to provide a product using the above character encoding method.

In order to achieve the above object, the present invention adopts the following technical solutions:

A character encoding method with error correction function, wherein the character encoding is composed of data characters and error correcting characters, and includes the following steps:

(1) generating a set of data characters according to the character encoding table;

(2) generating an error correction character corresponding to the data character using an error correction algorithm according to a preset error correction rate;

(3) A character code having an error correction function is composed of data characters and corresponding error correction characters.

Preferably, the character encoding table used in the step (1) includes a plurality of numbers and a plurality of letters, and different numbers and different letters represent different values.

Preferably, the character encoding table does not include a letter whose shape is close to any of the numerical shapes.

Preferably, in the step (2), the error correction algorithm used is the Reed Solomon algorithm.

Preferably, the length of the data character is N, and the number of characters to be error-corrected is T, and the length of the character encoding is N+2T, and 2T is the number of error-correcting characters; The error correction rate is T/(N+2T).

Preferably, the character code consists of numbers and/or letters.

A character decoding method with error correction function, wherein the character encoding is composed of data characters and error correction characters, including the following steps:

(1) Obtain an image of character encoding and identify the number or letter corresponding to each character;

(2) dividing the identified entire set of characters into data characters and error correction characters according to an error correction rate and an error correction algorithm;

(3) performing error correction processing on the character encoding using a corresponding error correction algorithm; if the error correction is successful, outputting the data character after deleting the error correction character; if the error correction fails, the entire data character encoding is eliminated.

Preferably, the error correction algorithm used in the step (2) and the step (3) is the Reed Solomon algorithm.

Preferably, in the step (2), the length N of the data character and the length of the error correction character are calculated according to the error correction rate T/(N+2T) and the length of the character encoding (N+2T). 2T, then divide; where T is the number of characters required for error correction.

A product provided with a character encoding composed of data characters and error correcting characters and obtained by the above encoding method.

Preferably, the product is a bottle cap, and the character code is disposed on the cap The top surface, the character codes set on different caps are different.

Preferably, a two-dimensional code is disposed inside the bottle cap, and the two-dimensional code disposed on each cap is different, and the two-dimensional code is associated with the character code in a database.

The character encoding method and the character decoding method provided by the present invention are used for encoding and decoding a character encoding having an error correction function. This character encoding with error correction function consists of data characters and their corresponding error correction characters. The use of such character encoding to mark the product can effectively solve the problem that the character encoding including only the data characters in the prior art is easily misunderstood. The error correction algorithm performs error correction processing on the character encoding, and can determine whether there is an error in the recognition of the character encoding, and correct the character encoding in which the number of the wrong characters is less than or equal to the number of error correction characters corresponding to the error correction rate; False failed character encoding. Although the number of error characters exceeds the number of error correction characters corresponding to the error correction rate, the erroneous characters in the character encoding cannot be completely corrected, but the recognition of the character encoding can also be determined by the error correction algorithm. The error is guaranteed to be 100% rejected. Therefore, using this character encoding, it can be ensured that only correctly recognized data characters can be stored in the system, and 100% of the incorrectly recognized data characters are rejected, and there is no misreading data in the system.

DRAWINGS

1 is an example of character encoding provided by the present invention;

2 is a flow chart of a coding process of character encoding provided by the present invention;

3 is a flow chart of a decoding process of character encoding provided by the present invention;

Figure 4 is a front elevational view showing a bottle cap provided with a character code;

Figure 5 is a rear perspective view of the cap of Figure 4;

detailed description

The technical content of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

The character encoding method and the character decoding method provided by the present invention are used for encoding and decoding a character encoding having an error correction function. As shown in FIG. 1, such a character code having an error correction function is composed of data characters and their corresponding error correction characters. In the decoding process, by performing error correction processing on the entire character encoding, it can be determined whether the error correction of the character encoding is successful, thereby ensuring that only correctly decoded data characters can enter the database, and the error correction data (ie, recognition error and cannot The error corrected data) was rejected by 100%.

In the following, the coding principle of the character encoding provided by the present invention will first be introduced. Ben The coding principle of the invention is as follows: The character code having the error correction function is composed of two parts: a data character and an error correction character. It is assumed that the error correction processing is performed by using the Reed Solomon algorithm, the number of data characters is N, and the number of characters to be error-corrected is T, the length of the character encoding is N+2T, and 2T is the number of error-correcting characters. The corresponding error correction rate is T/(N+2T). For example, in the character encoding shown in FIG. 1, assuming that two erroneous characters can be corrected in eight data characters, the length of the actual character encoding is 8 + 2 × 2 = 12, that is, a 12-bit character. For encoding, the length of the valid data character is 8, the length of the error correcting character is 4, the number of characters capable of error correction is 2, and the error correction rate is 2/12 = 16%. Since the recognition rate of OCR can reach 90%, plus 16% error correction rate, the system can basically guarantee 96% correct recognition rate. Most importantly, when using this character encoding with error correction, 100% misunderstanding can be found. Similarly, when using other lengths of data characters and error correction characters to form a character code, although the error correction rate and the correct recognition rate are different, the misreading can also be found 100%.

The error correction character is data obtained by processing the data characters using an error correction algorithm. According to the actual needs, the number of characters that need to be corrected can be adjusted. If the error correction is 1 character or 3 characters, the length of the error correction character is also adjusted to 2 or 6.

Any type of error correction algorithm can be used in the present invention as an error correction algorithm for the character encoding. For example, the commonly used Reed Solomon algorithm is used for error checking and error correction. For a detailed introduction to the Reed-Solomon algorithm, please refer to the paper "Reed-Solomon Codec Principle and FPGA Implementation" by Zhou Zhu and Wang Rui (published in "Avionics Technology", No. 2, 2009), which is not specific here. I will go into details. Of course, other error correction algorithms can be selected as needed.

The character code with error correction function provided by the present invention may be composed of pure letters or pure numbers, or may be composed of numbers and letters. The length of the character code can be freely defined as needed, generally 8 or more.

The character encoding method and the character decoding method provided by the present invention are described below by taking the character encoding table shown in Table 1 as an example.

The character encoding table shown in Table 1 is composed of 10 numbers and 22 uppercase letters, and includes 32 characters in total, and each character corresponds to a 5-digit binary number. When the character encoding includes 12 characters, the corresponding binary encoding length is 60 bits. In order to reduce misreading during the decoding process, in the character encoding table shown in Table 1, the letters B, I, O, and Q which are similar in shape to the easy-to-read error numbers 8, 1, 0 are not used. 32 characters after removing the above four letters The difference in shape is large, and the OCR device is easy to recognize and reduces the error rate.

Table 1 Character Code Table

When using the above character encoding table to generate data characters with 8 characters, the actual data maximum value is 32 ⁸ =1099511627776. If the data length is not enough, the characters can be lengthened to 9, 10, etc., and the longer data length means the error correction rate. The reduction, for example, the data length is 16, and the error correction data is 4, and the error correction rate is 2/20 = 10%.

Table 1 is a character encoding table corresponding to 32 values. In actual use, other character encoding tables corresponding to a plurality of values composed of only numbers, only letters, or partial digits and partial letters may be selected as needed. For example, you can use common hexadecimal values for encoding.

As shown in FIG. 2, the character encoding method with error correction function provided by the present invention includes the following steps: (1) setting an error correction rate of the error correction character generating device; that is, setting the number of characters that can be corrected, such as Set one error correction character; (2) generate a set of data characters according to the character encoding table; the character encoding table can be arbitrarily selected according to requirements; (3) according to the set error correction rate, the error correction character generating device generates and The error correction character corresponding to the data character; when processed by the Reed Solomon algorithm, the number of error correction characters generated is twice the number of error correction characters set; (4) the data characters and steps generated by the step (2) ( 3) The generated error correction characters constitute a character code with an error correction function. Then you can set the character encoding in production. On the product, used to mark the product, the character code can be set on the product in one or more lines.

When the error correction rate set in the step (1) is 1 character, the error correction is performed using the Reed Solomon algorithm, and if the length of the data character is N, the error correction rate is 1/(N + 2 × 1). At this time, in the decoding process, if there is a character reading error, it can be corrected. If there is a 2 character reading error, it can only be judged that the data character has an error and cannot be corrected. In the case where the number of error corrections is constant, the more the number of data characters, the lower the error correction rate.

Taking the character encoding shown in FIG. 1 as an example, it is assumed that the data character ACDE1234 is generated by using the character encoding table shown in Table 1; the number of characters requiring error correction is set to 2; and the error correction character generating device is used to generate and data using the error correction algorithm. The four error correction characters corresponding to the character, such as EGH2 (this is an example, the actual error correction value needs to be obtained according to a specific error correction algorithm); combining the data characters and the error correction characters to form an error correction function Character encoding, such as ACDE1234EGH2.

As shown in FIG. 3, the character decoding method with error correction function provided by the present invention is used for decoding the character encoding generated by the above encoding method; and the following steps are included: (1) acquiring a character encoded image and identifying each The number or letter corresponding to each character; (2) According to the error correction rate and the error correction algorithm, the entire set of characters recognized is divided into data characters and error correction characters; (3) the error correction decoding device uses the corresponding error correction algorithm Error correction processing is performed on the character encoding; if the error correction is successful, the data character is output; if the error correction fails, the entire data character encoding is eliminated.

Specifically, in the step (1), the OCR device can be used to read the character code having the error correction function; when the OCR device is used for reading, the recognition rate can be more than 90%. In step (2), if the Reed Solomon algorithm is used for processing in the encoding process, the length of the data character N may be first calculated according to the error correction rate T/(N+2T) and the length of the character encoding (N+2T). And the length of the error correction character is 2T, and then the character division is performed; wherein T is the number of characters required for error correction. When other error correction algorithms are used, the length of the data characters and the length of the error correction characters can also be first calculated according to the error correction algorithm, the error correction rate, and the length of the character encoding, and then the character division is performed.

The error correction decoding apparatus used in the above step (3) is a device capable of discriminating an error and correcting an error after inputting a character code, and when the number of error characters is less than or equal to the number of error correction characters set, the error is read. Characters can be corrected, error correction is successful; otherwise, only recognition errors can be found Errors can not correct the wrong characters, error correction failed. In step (3), the error correction algorithm used by the error correction decoding device should be identical to the error correction algorithm used to generate the character encoding.

Since the character encoding provided by the present invention is generally set for use on a product, in step (3), the process of rejecting the entire set of data character encoding can also be understood as a process of culling a product provided with the data character encoding. In this case, the rejecting device used in the step (3) means a device that rejects the defective product. Of course, the data characters that failed the error correction are not saved in the database.

In addition, in the step (3), in the process of correcting the character encoding by using the corresponding error correction algorithm, the error correction decoding device may further include a process of converting the character data and the error correction data into corresponding values using the character encoding table. Then perform error correction operations.

It should be noted that the above-mentioned error correction character generating device, error correction decoding device, culling setting, and the like may be implemented by an FPGA or firmware, or may be implemented by a dedicated software program running in a general-purpose microprocessor. The specific implementation manners are conventional technical means that can be grasped by those skilled in the art, and are not specifically described herein.

Next, the bottle cap shown in Figs. 4 and 5 will be taken as an example to introduce a product provided with a character code having an error correction function. At the same time, the bottle cap is taken as an example to introduce the production system for producing the cap.

In order to prevent stockpiling, many products are now equipped with character codes containing data characters. When the company produces products, the character codes with anti-smashing functions are set on the products by means of printers, laser machines and the like. Then, before the product leaves the factory, the OCR technology is used to identify and store the recognized character code in the database for subsequent information inquiry. Due to the limitations of OCR technology, when OCR devices read character encoding, there may be cases where characters cannot be read and misread. Among them, when the character cannot be read and cannot be recognized, the product can be directly culled; however, it cannot be discriminated for misreading. For example, when B is misread as 8, the OCR technology cannot determine whether misreading has occurred. Moreover, in the sales process, misreading is also prone to occur; in the prior art, the usual method for misreading is to compare the read data with the database, but due to the large amount of data and slow query speed, the actual It is difficult to operate in the application.

To this end, the present invention provides such a product provided with the above character encoding, wherein the character encoding is composed of data characters and their corresponding error correcting characters. As described in the above encoding method, after inputting data characters, the error correction character generating means generates an error based on the error correction rate and the error correction algorithm. Wrong character. The character code is then composed of data characters and error correction characters and placed on the product. The corresponding values of the character codes set on different products are different and can be used as product codes for inquiring in the production and sales process.

In addition, with the popularity of smart phones, QR code recognition software has become a must-have software for mobile phones, and a large number of mobile phone applications contain two-dimensional code recognition functions. In order to attract users to consume products, enterprises have set the lottery two-dimensional code on many products. In order to prevent users from scanning the QR code before purchase, the integral lottery two-dimensional code is usually hidden in the product packaging, and only after the purchase is destroyed by the packaging. QR code. Therefore, in the database, the character encoding needs to be associated with the two-dimensional code, so that the manufacturer can judge the destination of the corresponding two-dimensional code by reading the character encoding set on the product. By setting the attributes of the participating QR codes in the back-end database, etc., the product marketing activities can be guaranteed to proceed normally.

The above-mentioned product and production system provided with character encoding and two-dimensional code with error correction function are particularly suitable for use in various caps such as crown caps and the like. Before the factory, the two-dimensional code and character code set on the bottle cap are identified by the visual recognition device and the two are stored in the background database to ensure the normal sale of the bottle cap.

Specifically, the crown cap shown in FIGS. 4 and 5 includes a cover body 11. A two-dimensional code 14, a glue layer, and an edible sealant pad 15 are sequentially disposed downward on the inner side of the lid body 11. The edible sealant pad 15 is disposed on the surface of the two-dimensional code 14 and the adhesive layer. The edible sealing rubber pad 15 is made of edible silica gel material, and the two-dimensional code 14 can be insulated from the liquid in the bottle by setting the edible sealing rubber pad 15 to prevent the liquid from dissolving and destroying the two-dimensional code 14, and avoiding the two-dimensional code 14 from causing liquid in the bottle. Pollution.

A bottom oil layer and a surface oil layer are disposed in this order on the top side of the lid body 11. Wherein, after removing a part of the area of the surface oil layer by a process such as laser engraving, the exposed visible oil layer forms a character code 13. By changing the path of laser engraving, character codes 13 corresponding to different values can be formed. The character code 13 is composed of data characters and error correction characters, and the character code 13 is obtained by the above encoding method. The character code 13 can be composed of letters and/or numbers, has a short length, is easy to recognize in the production process, and has a fast recognition speed. The character code 13 can be arranged in one or more lines. In order not to affect the appearance of the product, the base oil layer may be a fluorescent ink or a transparent ink. Further, a printed pattern may be disposed between the base oil layer and the oil layer.

For the above cap, the two-dimensional code 14 and the character code 13 are identified and associated in the background, having a unique correspondence, thereby constituting two identification codes of the cap. Among them, set in different The two-dimensional codes on the caps are different from each other, and the character codes set on different caps are different from each other, and the two identification codes respectively become the identity of the cap. The QR code 14 can be used for activities such as points, lotteries, and the like after the consumer destroys the package. By arranging the two-dimensional code information layer 14 on the inner side (i.e., the inner surface) of the cover 1, it is possible to prevent the participating two-dimensional code from being damaged or counterfeited during the sales process. The character code 13 provided on the top side of the cover can be used for anti-smuggling during production and sales.

The product is marked by the use of such character encoding with error correction. In the identification process, if the error correction of the character encoding set on the product is successful, the character data of the product is stored in the database and stored in association with the two-dimensional code set on the back of the product. If the error correction of the character encoding set on the product is unsuccessful, the above-mentioned character encoding recognition error is judged, thereby eliminating the corresponding product, and deleting the relevant data in the database, thereby ensuring the character encoding on the product entering the sales link. 100% identifiable. In the sales process, the manufacturer or the dealer can know the whereabouts of the product by identifying the character code on the outside of the bottle cap, and carry out the anti-smuggling inquiry; and, by identifying the character code on the outside of the cap, the second side of the cap can be determined. Dimensional code, so that by adjusting the winning range of the QR code, the marketing activities can be guaranteed to proceed normally. By setting a different two-dimensional code on the inside of each cap, each bottle of beverage or beer has a unique identity, and the consumer can participate in the manufacturer's activities by acquiring the two-dimensional code.

In summary, the present invention provides a character encoding method and a character decoding method for encoding and decoding a character encoding having an error correction function. When such a character encoding with error correction function is set on the product, it is possible to judge whether there is an error in the recognition of the character encoding through the error correction algorithm, and it can be ensured that only the data characters with successful error correction can be stored in the system, and the data character 100 that fails the error correction is ensured. % is culled to ensure that there are no misreading data in the data system and to avoid data misreading during the sales process.

The character encoding method, the character decoding method, and the product thereof having the error correction function provided by the present invention have been described in detail above. Any obvious modifications made to it by those skilled in the art without departing from the spirit of the invention should fall within the scope of the invention.

Claims

A character encoding method with error correction function, comprising the following steps:

(1) generating a set of data characters according to the character encoding table;

(2) generating an error correction character corresponding to the data character using an error correction algorithm according to a preset error correction rate;

(3) A character code having an error correction function is composed of data characters and corresponding error correction characters.
A character encoding method according to claim 1, wherein:

The character encoding table used in the step (1) includes a plurality of numbers and a plurality of letters, and different numbers and different letters respectively represent different numerical values.
A character encoding method according to claim 2, wherein:

Letters whose shapes are close to any of the numerical shapes are not included in the character encoding table.
A character encoding method according to claim 1, wherein:

In the step (2), the error correction algorithm used is the Reed Solomon algorithm.
A character encoding method according to claim 4, wherein:

The length of the data character is N, the number of characters to be error-corrected is T, and the length of the character encoding is N+2T, and 2T is the number of error-correcting characters; the error correction rate corresponding to the character encoding is T/(N+2T).
A character encoding method according to claim 1, wherein:

The character code consists of numbers and/or letters.
A character decoding method with error correction function, comprising the following steps:

(1) Obtain an image of character encoding and identify the number or letter corresponding to each character;

(2) According to the error correction algorithm and the error correction rate, the identified entire set of characters are divided into data characters and error correction characters;

(3) performing error correction processing on the character encoding using a corresponding error correction algorithm; if the error correction is successful, outputting the data character after deleting the error correction character; if the error correction fails, the entire data character encoding is eliminated.
A character decoding method according to claim 7, wherein:

The error correction algorithm used in the step (2) and the step (3) is the Reed Solomon algorithm.
A character decoding method according to claim 8, wherein:

In the step (2), the length N of the data character and the length 2T of the error correction character are calculated according to the error correction rate T/(N+2T) and the length of the character encoding (N+2T), and then divided. Where T is the number of characters required for error correction.
A character decoding method according to claim 7, wherein:

In the step (3), in the process of performing error correction processing on the character encoding using an error correction algorithm, the method further includes converting a data character and an error correction character into a corresponding numerical value by using a character encoding table; wherein the character encoding The table includes a plurality of numbers and a plurality of letters, and different numbers and different letters represent different values.
A character decoding method according to claim 10, wherein:

Letters whose shapes are close to any of the numerical shapes are not included in the character encoding table.
A product provided with a character code, characterized in that the character code is composed of a data character and an error correction character, and is obtained by the character encoding method according to any one of claims 1 to 6.
The product of claim 12 wherein:

The product is a bottle cap, and the character code is disposed on a top surface of the cap, and the character codes provided on different caps are different.
The product of claim 13 wherein:

A two-dimensional code is disposed inside the cap, and a two-dimensional code disposed on each cap is different; the two-dimensional code is associated with the character encoding in a database.