WO2019080818A1

WO2019080818A1 - Bar code recognition based book positioning method, electronic device and storage medium

Info

Publication number: WO2019080818A1
Application number: PCT/CN2018/111317
Authority: WO
Inventors: 邓立邦
Original assignee: 广东数相智能科技有限公司
Priority date: 2017-10-23
Filing date: 2018-10-23
Publication date: 2019-05-02
Also published as: CN108009602A; CN108009602B

Abstract

A bar code recognition based book positioning method, an electronic device and a storage medium. The bar code recognition based book positioning method comprises the following steps: acquiring image information of a bookshelf and generating a position number of a camera generating the image information, the image information comprising at least one book image, and the book image comprising a bar code area (S1); according to the image information and the position of the bar code area in the image information, obtaining a row and column position, on the bookshelf, of a book corresponding to the bar code area (S3); and obtaining a storage position of the corresponding book according to the row and column position of the book corresponding to the bar code area and the position number of the camera (S4). In the bar code recognition based book positioning method, by means of a bar code corresponding to a book and the position of a corresponding camera acquiring the bar code, a storage position corresponding to the book is obtained, improving the efficiency of book positioning.

Description

Book positioning method based on barcode recognition, electronic device and storage medium

Technical field

The invention relates to the field of library management, in particular to a book positioning method, an electronic device and a storage medium based on barcode recognition.

Background technique

At present, librarians spend a lot of energy every day on borrowing, returning, degaussing, counting, loading, warehousing, sorting, and wrapping books. Their work is trivial and complicated, and the workload is huge, which requires a lot of time, energy and physical strength. Due to the borrowing personnel in the process of searching for books, the homing of the books after viewing is often quite random. On various bookshelves, there are often problems in that books are not placed in accordance with the rules, and other borrowers cannot find books that need to be borrowed. Librarians also need to spend a lot of time every day to check and sort out such misplaced books, which is heavy and time consuming. And because the inspection work needs to rely on the experience and memory of the librarian to remember the correct storage location of all books in order to carry out rapid investigation and homing, it is obviously very difficult and prone to omissions. In the process of returning books, librarians also need to keep in mind the general classification and storage location of books that are often borrowed. For example, look for the bookshelf on the spine of the book, find out which layer should be placed on the bookshelf, and finally see if there is any If you have the same book, you need to put it together. If you are looking for the same kind of book, put it together. If you are not familiar with the book storage classification and location, it will take 7-8 minutes to return the book. Sorting and sorting, and manual operation by memory is inevitable, and work efficiency is low. It is often necessary to work overtime when closing, which takes a lot of time and manpower. Therefore, how to improve work efficiency, save errors and organize books, store homing books, and reduce the workload of librarians becomes a problem that we need to think about.

Moreover, the identification of the existing bar code code relies on professional equipment. If the requirements of the existing camera are relatively high, the recognition accuracy cannot be large.

Summary of the invention

In order to overcome the deficiencies of the prior art, one of the objects of the present invention is to provide a book positioning method based on barcode recognition, which can solve the technical problem of automatically storing books.

Another object of the present invention is to provide an electronic device capable of solving the technical problem of book positioning.

A third object of the present invention is to provide a computer readable storage medium capable of solving the technical problem of storing a book location.

The fourth object of the present invention is a book positioning method based on barcode recognition, which can solve the technical problem of automatically checking books.

One of the objects of the present invention is achieved by the following technical solutions:

A book positioning method based on barcode recognition includes the following steps:

An information acquisition step of: acquiring image information of a bookshelf and a position number of a camera that generates the image information, the image information including at least one book image, the book image including a barcode area;

Position acquisition step: according to the image information and the position of the barcode area in the image information, the position of the book corresponding to the barcode area on the shelf;

Result acquisition step: according to the position of the row and the position of the book corresponding to the barcode area and the position number of the camera to obtain the storage position of the corresponding book.

Further, after the information obtaining step, the method further includes a book information obtaining step of: extracting color values of all the color blocks in the barcode area, and obtaining character information corresponding to each color block in the barcode area according to the barcode encoding rule, thereby obtaining corresponding book information, The color blocks in the barcode area are sequentially arranged in order, and the character information and the color value are in one-to-one correspondence to form a barcode encoding rule.

Further, after the information acquiring step, an image pre-processing step is further included: pre-processing the acquired current image information, the pre-processing including removing noise and interference, image cutting, and image normalization.

Further, after the information obtaining step, the step of identifying the comparison is further performed: comparing the current image information with the barcode recognition model to obtain a barcode area.

Further, after the identification comparison step, a 3D correction step is further included: 3D correction is performed on the barcode area to adjust the barcode area to a rectangular barcode area.

Further, the process of establishing the barcode recognition model includes the following sub-steps:

Model pre-processing step: pre-processing the acquired training barcode image, the pre-processing including removing noise and interference, image cutting and image normalization;

Feature extraction step: performing feature extraction on the pre-processed training barcode image to obtain a barcode training set;

Model establishment step: extracting a standard template from the barcode training set to establish a barcode recognition model.

Further, after the character obtaining step, a character verification step is further included: the character information is verified according to a verification rule to confirm whether it is correct.

Further, after the result obtaining step, the method further includes: a comparison step of comparing the classification information in the character information with the book classification area corresponding to the position number of the camera, and determining whether the book corresponding to the character information belongs to the book classification area If no, send a reminder to the administrator.

Further, after the result obtaining step, the method further includes a position comparison step of comparing the storage location of the obtained book with the preset storage location of the book, and if not, sending the reminder information to the administrator.

The second object of the present invention is achieved by the following technical solutions:

An electronic device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor executing the program to implement the bar code based on any one of the objects of the present invention Recognized book location method.

The third object of the present invention is achieved by the following technical solutions:

A computer readable storage medium having stored thereon a computer program, wherein the computer program is executed by a processor to implement a bar code recognition based book positioning method according to any one of the objects of the present invention.

The fourth object of the present invention is achieved by the following technical solutions:

Information acquiring step: acquiring image information of a bookshelf, the image information including a bookshelf number and at least one book image, the book image including a barcode area;

Result obtaining step: according to the position of the row and the book number corresponding to the book corresponding to the barcode area, the storage position of the corresponding book is obtained.

Compared with the prior art, the beneficial effects of the present invention are:

The barcode identification-based book positioning method of the invention obtains the storage location corresponding to the book by the barcode corresponding to the book and the position of the camera corresponding to the barcode; and improves the efficiency of the book positioning.

DRAWINGS

1 is a flow chart of a method for positioning a book based on barcode recognition according to the present invention.

Detailed ways

The present invention will be further described in conjunction with the accompanying drawings and specific embodiments. It should be noted that, without conflict, the embodiments described below may be arbitrarily combined to form a new embodiment. .

The invention consists of: a color bar code, an acquisition module, an identification module, a data module, a judgment module, an alert module, and a book query screen, and the functions of each part are as follows:

Color bar code: A color bar code that is encoded with a specific color value corresponding to different strips and null characters. The color bar code is used to paste on the spine of the side of the book, and each strip and space is represented by a uniform color block of the same shape. In order to allow the camera to shoot and identify it in a range of distances. The barcode is based on the book category, the library, the storage area, the bookshelf number, and the storage layer on the shelf. The encoding rules are used, and 10 different HSB color values are used to correspond to 10 digits of 0-9. The start character and the check character are combined by using the corresponding character color value color block + black color block with directivity, so that the computer can recognize and judge according to the character appearance, the shape and the color combination feature, and realize the scan code verification.

Acquisition module: a camera set for multiple sets of bookshelves placed in the corresponding areas in the library. Each set of cameras is numbered according to each area and the booked book number. The position of the camera is supported to be fully captured at a certain distance. Side image of the book on the shelf. The side of the book arranged on the bookshelf of each area is photographed by the camera, and the color bar code image attached to the side spine of all the books on the bookshelf is obtained and sent to the identification module.

The identification module: receiving the side image of the book sent by the acquisition module, and dividing the side of each book in the captured image to compare the pre-established barcode recognition model, and extracting the barcode image area corresponding to the paste on the side spine of the book, according to the coding of the barcode The rule identifies the barcode image.

Data module: A barcode database that stores the coding rules and content information of the color barcode. At the same time, the book information corresponding to each color bar code is saved, the books are stored in the bookshelf and the storage location information on the bookshelf, the pasted bar code information, and the like. By accessing the pasted bar code information, when the color bar code is set, it can be entered in one-to-one correspondence with the bar code information already existing on the book, so that the existing book information system can be accessed, and the existing book information system can be upgraded and updated more conveniently and quickly. It interfaces with the system data of the present invention without requiring a complete replacement update of the library's existing information system. If the system is completely used, there is no need to access the existing system of the library, and the data module can also establish information such as book introduction information, classified retrieval information, borrowing information and other book management requirements.

Judgment module: according to the information identification result of the color bar code, the number information after the single book is cut, the book storage book shelf number and the storage location information in the data module are retrieved, and the correctness of the book storage is judged, and if the book stored in error is found, The title of the book, the current storage location and the correct storage location information are sent to the reminder module.

The reminder module: displays the book information stored in error according to the judgment result, and performs a voice broadcast to remind the librarian or the book borrower to organize the book in time.

Book inquiry screen: An electronic touch screen display that is arranged in various areas of the library to facilitate borrowers or librarians to inquire about books. The book inquiry screen is provided with a camera and a voice playing module, and a book positioning system based on a camera and a barcode is also connected. Borrowers and librarians can use the book inquiry screen to scan and identify the barcode on the book, obtain the storage location information of the book, and query the bookshelf and storage location for books that do not need to be borrowed or need to be returned. Borrowing book information.

An electronic device includes a memory, a processor, and a computer program stored on the memory and operable on the processor. As shown in FIG. 1, the processor performs the following steps when executing the program:

S1: acquiring image information of the bookshelf and a position number of the camera that generates the image information, the image information including at least one book image, the book image including a barcode area; in addition to the above manner, another way is to set the camera to shoot the bookshelf When the image information is included, the numbered area of the bookshelf is included to obtain the number of the bookshelf, and the text information contained in the image of the bookshelf number area is recognized by the character recognition technology to obtain the bookshelf number, so as to locate the book shelf where the book is located;

S11: Perform preprocessing on the obtained current image information, including preprocessing noise and interference, image cutting, and image normalization; and improving the server by a preprocessing process such as noise and interference, image cutting, and normalization. The performance of image processing recognition; preprocessing the obtained side image of the book, and obtaining the book image of each layer of the bookshelf in the image information through color clustering and color block segmentation. Edge detection is used to segment images of multiple books contained in the image according to the pixel arrangement features at the edge of each book, and to obtain images of multiple books on the side of the book.

S12: Comparing the current image information with the barcode recognition model to obtain a barcode area; the process of establishing the barcode recognition model includes the following substeps:

Model preprocessing step: preprocessing the obtained training barcode image, the preprocessing includes removing noise and interference, image cutting and image normalization; and preprocessing processes such as noise and interference, image cutting, normalization, etc. Improve the performance of the server for image processing recognition;

Feature extraction step: performing feature extraction on the pre-processed training barcode image to obtain a barcode training set; performing feature extraction on the preprocessed color barcode image, according to the color density of the color barcode image, and the pattern arrangement characteristics The barcode image is divided into 25 square regions of 5*5, and the ratio of the number of points in each square to the total number of points of the color barcode image is calculated to obtain a 25-dimensional feature vector;

Model establishment step: extracting a standard template from the barcode training set to establish a barcode recognition model. Through repeated recognition training, the standard template is extracted from the color bar code images of the training set, and the standard feature library is established to complete the establishment of the bar code recognition model. Each type of color bar code has hundreds of standard templates. Because the bar code has directionality, the bar code image needs to be correctly arranged in the starting direction and then provided to the machine for learning, so that the later machine recognizes and processes the bar code image. It can be automatically judged and the starting position of the barcode is calibrated. After preprocessing and feature extraction, the feature vectors of each color bar code of the training set are stored in the file, and the correct value of each bar code needs to be specified during training.

S13: performing 3D correction on the barcode area to adjust the barcode area to a rectangular barcode area; determining, by the comparison recognition model, the color barcode image area included in the side picture of the single book, positioning, 3D correction, and barcode content on the color barcode image Identification.

Bar code positioning: feature extraction of the side images of the divided books, find out the colors, patterns, shapes and other attributes that constitute the distinctive features of the color bar codes, calculate their geometric feature quantities, and form the feature vectors describing the color bar codes through these feature quantities. And comparing with the feature vector of the barcode recognition model established in the system, when the similarity exceeds 80%, the region can be determined as a color barcode region, and the color barcode region is positioned according to the judgment result.

3D correction: Due to the shooting angle, the shooting angle of the barcode on the spine, the tilt angle of the book, etc., the color barcode image usually has a certain degree of image distortion or angular rotation reversed, so 3D of the color barcode image is required. Correction. Through spatial mapping, the four points of the upper, lower, left and right corners of the color barcode image are fixed and then projected, and the color barcode image is rotated and twisted, so that the barcode image is arranged according to the starting direction of the barcode. Become a standard rectangular area and get a color bar code image of standard shape size.

S2: extracting color values of all the color blocks in the barcode area, and obtaining character information corresponding to each color block in the barcode area according to the barcode encoding rule, thereby obtaining corresponding book information, wherein the color blocks in the barcode area are sequentially arranged in order, The character information and the color value are in one-to-one correspondence to form a bar code encoding rule; since the black and white bar code is realized by the interval between the bar spaces, the required precision is high, and if the distance is far away, there is no way to identify the gap between the bar spaces. The color bar code-based color bar code identifies the information by the color value of the color block, so if the camera is far away, the color can be recognized as long as the color in the middle of the color block can be extracted; therefore, the color bar code of the present invention can realize large Technical issues of area identification.

Bar code content identification. The bar code information is identified on the color bar code image of each book after 3D correction processing according to the color bar code encoding rule pre-stored in the system. First, it is necessary to judge the color value of each bar. The color bar code uses various color blocks of equal width to represent each bar and space character. After taking the average value of each bar and block of the color bar code, extract each bar. The HSB color value information of the block sequentially determines the character information corresponding to each color value according to the encoding rule.

S21: Check the character information according to a verification rule to confirm whether it is correct. The check rule is also a check code setting rule. Since each color bar code has a check code, the check code can be obtained according to the previous data, so that it can not only display by color, but also It can be obtained by calculation. When the test is performed, the corresponding character of the check code can be calculated to compare with the character corresponding to the recognized color to achieve the purpose of verifying the character information;

S3: according to the layer corresponding to the book shelf where the book is located in the divided image information, and the position of the barcode area in the layered image information of the divided bookshelf, the storage layer position of the book corresponding to the barcode area on the bookshelf; since step S1 has been completed The segmentation of the book image at each layer of the bookshelf in the book image and the segmentation of each book stored on each shelf. Therefore, each book image of each layer of the segment is sequentially numbered according to a certain positional arrangement rule. For example, "camera number - bookshelf layer - column position" is used to determine whether the storage location of the book is correct.

S4: According to the position of the row and the position of the book corresponding to the barcode area and the position number of the camera, the storage position of the corresponding book is obtained.

The above scheme is only for book positioning, in addition to book positioning, it is also possible to identify the corresponding misplaced book by position comparison; the main methods are as follows:

S4a: The color bar code character is encoded according to the information of each aspect of the book, and according to the character recognition result and the encoding rule of the color bar code, the classification information in the book information is compared with the book classification area corresponding to the position number of the camera, and the judgment is made. Whether the book corresponding to the book information belongs to the book classification area, if not, sending the reminder information to the administrator; by extracting the classification information in the book information, that is, the belonging area of the category corresponding to the book, if the classified area Remind the administrator if it is inconsistent with the current location;

S4b: Compare the storage location of the obtained book with the preset storage location of the book, and if not, send the reminder information to the administrator. In addition to the above identification methods, there is also a way to directly identify all the character information and extract corresponding books different from the classification information in most of the character information, thereby reminding the administrator that the process can be every Automatic identification reminders are made every hour or every four hours, or they can be identified one hour before closing.

The invention provides a book positioning system and device based on a camera and a barcode, which can identify the color barcodes in batches, accurately and quickly by the camera, thereby quickly positioning the books of the library without using a professional barcode scanning device. The scanning efficiency is improved, and the problems of slow manual operation, error in manual memory, and low efficiency are solved. By arranging a reasonable number of cameras for bookshelf in various areas of the library, monitoring and analyzing the books stored on the bookshelf, automatically checking and prompting the books stored in error, reducing the time and effort of manual investigation, improving work efficiency, and having a certain economy. Benefits and market application value. At the same time, through the inquiry screens arranged in the respective deposit areas, the borrower can query the storage locations of the books that do not need to be borrowed in real time after consulting the books, so as to help them accurately return the books to the correct storage shelves. In order to improve the random storage of books, the workload of librarians has increased significantly.

The above embodiments are merely preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, and any insubstantial changes and substitutions made by those skilled in the art based on the present invention belong to the present invention. The scope of the claim.

Claims

A book positioning method based on barcode recognition, comprising the following steps:

An information acquisition step of: acquiring image information of a bookshelf and a position number of a camera that generates the image information, the image information including at least one book image, the book image including a barcode area;

Position acquisition step: according to the image information and the position of the barcode area in the image information, the position of the book corresponding to the barcode area on the shelf;

Result acquisition step: according to the position of the row and the position of the book corresponding to the barcode area and the position number of the camera to obtain the storage position of the corresponding book.
The bar code recognition-based book locating method according to claim 1, further comprising a book information obtaining step of: extracting color values of all color blocks in the bar code region, and obtaining a bar code according to the bar code encoding rule. The character information corresponding to each color block in the area, thereby obtaining corresponding book information, wherein the color blocks in the bar code area are sequentially arranged in order, and the character information and the color value are in one-to-one correspondence to form a bar code encoding rule.
The bar code recognition-based book locating method according to claim 1, further comprising an image pre-processing step of: pre-processing the acquired current image information, the pre-processing including removing noise after the information acquiring step And interference, image cutting, and image normalization.
The method for positioning a book based on bar code recognition according to claim 3, further comprising the following steps after the image preprocessing step:

Identifying the comparison step: comparing the current image information with the barcode recognition model to obtain a barcode area;

3D correction step: 3D correction is performed on the barcode area to adjust the barcode area to a rectangular barcode area.
The method for positioning a book based on bar code recognition according to claim 4, wherein the process of establishing the bar code recognition model comprises the following substeps:

Model pre-processing step: pre-processing the acquired training barcode image, the pre-processing including removing noise and interference, image cutting and image normalization;

Feature extraction step: performing feature extraction on the pre-processed training barcode image to obtain a barcode training set;

Model establishment step: extracting a standard template from the barcode training set to establish a barcode recognition model.
The bar code recognition-based book locating method according to claim 1, further comprising a character verification step after the character obtaining step: verifying the character information according to a verification rule to confirm whether it is correct.
The method for positioning a book based on bar code recognition according to claim 1, further comprising: after the result obtaining step, the step of comparing the positions of the obtained book and the preset storage position of the book; If they are inconsistent, send a reminder message to the administrator.
An electronic device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the processor executes the program to implement any one of claims 1-7 The book positioning method based on barcode recognition.
A computer readable storage medium having stored thereon a computer program, wherein the computer program is executed by a processor to implement a barcode recognition based book positioning method according to any one of claims 1-7.
A book positioning method based on barcode recognition, comprising the following steps:

Information acquiring step: acquiring image information of a bookshelf, the image information including a bookshelf number and at least one book image, the book image including a barcode area;

Position acquisition step: according to the image information and the position of the barcode area in the image information, the position of the book corresponding to the barcode area on the shelf;

Result obtaining step: according to the position of the row and the book number corresponding to the book corresponding to the barcode area, the storage position of the corresponding book is obtained.