CN109359705B - System and method for searching misplaced books - Google Patents

Abstract

The invention discloses a staggered book searching system and a searching method thereof. The method for searching the dislocation book specifically comprises the following steps: s1, shooting a picture of the front of the bookshelf and splicing the picture into a complete image P; s2, recognizing the image P and extracting book searching numbers on all the spine labels; s3, sequentially storing book searching numbers and coordinates of the books in the image P into a list A and a list POS; s4, scanning and analyzing the data of the list A, and detecting the staggered book searching number; s5, finding the coordinates of the misplaced book number in the list POS and marking the image P with a specific color; and S6, repeating S1 to S5 until the staggered book searching and marking of all the bookshelves are completed. The method has the advantages of simplicity, no need of changing the existing book equipment and low cost.

Description

System and method for searching misplaced books

Technical Field

The invention belongs to the technical field of library automation, and mainly relates to a system and a method for searching a misplaced book.

Background

The books collected in the modern society library are increased year by year, and book borrowing activities are increasingly frequent. When the reader returns the book, the administrator scans the bar codes of the books by using the code scanning equipment, puts the bar codes of the books in a book management system, and then puts the bar codes of the books at the appointed position of the bookshelf. However, the readers can not put back the books in place after reading the books in the reading room, the library staff need to regularly arrange the misplaced books, but the manual arrangement mode is low in efficiency and easy to miss, and the situation that the readers can not find the books on the corresponding bookshelves when inquiring about some books in the book management system and storing books in the library is frequently generated.

For the situation, the existing solution is to use the RFID radio frequency tag technology to locate the book, but there is still a certain obstacle in terms of use cost and system integration, and it is difficult to popularize on a large scale. Still other solutions have attempted to solve the book sorting problem, such as: chinese patent document CN106504443A discloses an automatic book arrangement device, which uses a walking vehicle body and a lifting mechanism and a moving mechanism arranged on the vehicle body to automatically arrange the book shelves. Chinese patent document CN107862488A discloses a crawler-type book arrangement system, in which a crawler scanning device is installed on a bookshelf to scan information of books on the bookshelf, and then the books are arranged through a condition component and a pushing component according to the information of the books on the bookshelf. The Chinese patent document CN107329469A discloses an independent robot disordered shelf book management system, wherein two-dimensional codes are coded and manufactured for each layer of a bookshelf, a scanning gun module on the robot scans codes to obtain bookshelf information, a bar code on a spine is scanned to obtain book information, and the robot scans the book bar code on the whole bookshelf to compare with original position information to judge whether books are wrongly shelved, so that disordered shelf book management is realized. The three existing schemes need to transform the bookshelf or use an automatic device to arrange the bookshelf, if the books need to be judged to be misplaced, the correct storage positions of the books need to be stored in the database in advance, the use and maintenance cost is high, and certain obstacles exist in popularization.

Disclosure of Invention

The main problem of book arrangement in a library is how to find out misplaced books, once the misplaced books are found, the subsequent arrangement work is completed manually and is not difficult; at present, a dislocation book searching method with relatively low price is still absent in the market. The invention provides a system and a method for searching a misplaced book, which are used for searching the misplaced book by utilizing image recognition and data processing technologies.

In order to solve the technical problems, the invention adopts the following technical means:

the utility model provides a dislocation books system of looking for, specifically includes parameter setting module, bookshelf scanning equipment, image concatenation module, book mark recognition module, dislocation books module of looking for and data report module.

The parameter setting module is used for setting the indoor plane layout, the number of the bookshelves, the positions of the bookshelves and the size of the bookshelves; the bookshelf scanning device is used for shooting bookshelf pictures in parts and uploading the shot pictures to the image splicing module; the image splicing module is used for splicing a plurality of bookshelf images shot and uploaded by the bookshelf scanning equipment into a complete image; the book mark identification module is used for identifying the book spine label image and converting the book spine label image into a book searching number in a text format; the staggered book searching module is used for analyzing the identified book searching number data and searching out staggered books in the identified book searching number data; the data report module stores the misplaced book information found by the misplaced book searching module and makes the misplaced book information into a report for a worker to check.

Furthermore, the bookshelf scanning device adopts automatic shooting equipment or manual shooting.

Further, the image splicing module adopts a Harris corner detection algorithm based on standard deviation to splice images.

A method for searching a misplaced book specifically comprises the following steps:

s1, shooting a plurality of book images on the front side of the bookshelf, uploading the images to the system, and splicing the images into a complete image P in sequence;

s2, performing text recognition on the image P, and extracting book searching numbers on all spine labels in the image P;

s3, sequentially storing book searching numbers and coordinates of corresponding book labels in the image P into a list A and a list POS;

s4, scanning and analyzing the book searching numbers stored in the list A, and detecting the displaced book searching numbers;

s5, finding out coordinate data corresponding to the book searching number of the misplaced book detected in the step S4 in the list POS, and marking the image P with a specific color according to the coordinate data;

and S6, repeating the steps S1 to S5 until the staggered book searching and marking of all the bookshelves are completed.

Further, step S3 sequentially stores the book index number and the coordinates of the corresponding book label in the image P in the list a and the list POS, specifically including the following steps:

s31, constructing a list A for sequentially storing book searching numbers of the books and constructing a list POS for storing horizontal coordinates and vertical coordinates of book labels in the image P;

s32, enabling T to be the current book and initially be the first book at the upper left corner of the image P;

s33, adding the book searching number of the T into the list A, and adding the coordinate of the T into the list POS;

s34, let T be the next book on the right side of the current book, if there is no book on the right side, let T be the leftmost book on the next layer in the image P, if the current book is the last book on the lower right corner in the image P, then the process is ended, otherwise go to step S33.

Further, the step S4 scans and analyzes the book search numbers stored in the list a, and detects the displaced book search numbers, which is specifically realized through the following steps:

s41, creating an array L with the same length as the list A, recording the longest increment subsequence length of each element in the list A taking the element as a tail element, and making L [1] equal to 1;

s42, creating an array P with the same length as the list A, recording the precursor elements of the elements in the list A in the longest increasing subsequence taking the elements as tail elements, and enabling P [1] to be NULL;

s43, defining a current subscript x, and making x equal to 2;

s44, detecting the index number of all elements before subscript x in list A, finding the longest increasing subsequence taking A [ x ] as the ending element, namely, for each i < x, calculating to obtain L [ x ] according to the following formula:

L[x]＝max{L(i)+1|A[i]<A[x]∧i<x} (1)

and recording P [ x ] as the previous element subscript i of the longest increment subsequence obtained when calculating L [ x ];

s45, making x equal to x +1, if x is less than or equal to | a |, turning to S44, otherwise, turning to S46;

s46, traversing the array L and finding the subscript k at the maximum value, and then tracing back the longest increasing subsequence taking A [ k ] as the tail element at P [ k ], and recording the longest increasing subsequence as LIS;

and S47, setting the misplaced book list B as all books of the list A except the LIS.

The basic idea of the method for searching the staggered books is as follows: the spine part of each book in the library is pasted with a label, a unique book searching number is printed on the label, and the books are always placed layer by layer in sequence according to the book searching number when being put on a shelf. If the placing sequence of the books on the bookshelf is all correct, the book searching numbers of the books inevitably form a monotone increasing sequence; on the contrary, if a misplaced book appears on the bookshelf, the monotone increasing property of the sequence is broken. Therefore, the misplaced books can be found as long as the wrong number in the book number sequence can be found.

The following advantages can be obtained by adopting the technical means:

according to the system and the method for searching the misplaced books, the misplaced books are searched through the image recognition and data analysis technology, the original books or library equipment does not need to be modified, the cost is low, the system and the method are not limited by the environment, and the self-adaptive capacity is strong. In addition, the misplaced books are retrieved according to the characteristic that the book searching number is increased progressively, the correct storage positions of the books are not required to be stored in a background database in advance, the operation complexity and the operation difficulty are greatly reduced, excessive maintenance is not required in subsequent use, and the method is easy to popularize and use.

Drawings

FIG. 1 is a schematic diagram of a system for searching a misaligned book according to the present invention.

FIG. 2 is a flowchart of a method for finding a misaligned book according to the present invention.

FIG. 3 is a schematic diagram of a book searching sequence in the method for searching a misaligned book according to the present invention.

FIG. 4 is a flow chart illustrating the operation of the present invention in analyzing data and searching for misaligned books.

Detailed Description

The technical scheme of the invention is further explained by combining the accompanying drawings as follows:

a staggered book searching system is shown in figure 1 and specifically comprises a parameter setting module, a bookshelf scanning device, an image splicing module, a book mark identification module, a staggered book searching module and a data report module.

The parameter setting module is used for setting parameters such as indoor plane layout, the number of the bookshelves, the positions of the bookshelves and the sizes of the bookshelves; the bookshelf scanning device is used for shooting bookshelf pictures in parts and uploading the shot pictures to the image splicing module, the shooting in parts is used for recording detail information of each book more clearly, and the bookshelf scanning device can select automatic shooting equipment and can also be shot and uploaded manually by workers; the image splicing module is used for splicing a plurality of bookshelf images shot and uploaded by the bookshelf scanning equipment into a clear and complete image; the book mark identification module is used for identifying the book spine label image and converting the book spine label image into a book searching number in a text format; the staggered book searching module is used for sorting and analyzing the recognized book searching number data and searching out staggered books therein; and the data report module stores the searched staggered book information and makes the staggered book information into a report for the staff to check.

A method for searching a misaligned book, as shown in fig. 2, specifically includes the following steps:

and S1, shooting a plurality of book images on the front side of the bookshelf, and splicing the book images into a complete image P. In the embodiment, a Harris corner detection algorithm based on standard deviation is adopted for feature detection, and then image splicing is performed. In the area with rich texture information, the Harris operator can extract a large number of useful feature points, and because the calculation process only involves the first derivative of the image, the extraction of the corner points is stable even if the image rotates, the gray level changes and the viewpoint changes. Each angular point of the image is represented by a matrix of 9 multiplied by 9, the Euclidean distance between each angular point descriptor and the corresponding angular point descriptor is calculated, all the angular points are roughly matched, then a pseudo-matching point pair is eliminated by using a random sampling consensus RANSAC method, a homography matrix is calculated, and an accurate matching point pair is obtained. And the images are fused by adopting a smooth transition method, so that the smooth transition between the images is realized, and the splicing trace is eliminated. And finally, recording the spliced complete image as P.

And S2, performing text recognition on the image P, and extracting the book searching number on the spine. The characters on the book label are printed characters, and the characters are more standard than handwriting characters, more regular in pixel distribution and beneficial to being identified by a statistical characteristic method.

In this embodiment, a recognition method based on statistical characteristics is adopted to recognize book tags, a template is obtained by statistical operation based on training sample pixels, and a character recognition result is obtained by matching the template and the book tags, and the specific steps are as follows:

and S21, preprocessing the image. Firstly, preprocessing an image, graying the image and then segmenting the image; then normalizing the segmented image, and correcting each character to be a uniform size by adopting a linear interpolation method; then, the OTSU algorithm is used for calculating a threshold value, and the image is binarized.

And S22, image recognition. Template training is carried out through statistics and distribution probability of pixel points of the alphabetic image, then classification and recognition are carried out through calculating the distance between a feature matrix of image characters and each character template, and an image recognition result of each book label is stored in one character string.

S3, storing the book searching number of each book and the coordinates of the label in the image P into a list A and a list POS according to the position sequence in the bookshelf respectively, and the specific steps are as follows:

s31, constructing a list A for storing the book searching number of each book; constructing a list POS for storing the abscissa and ordinate of each label in the image P;

s32, enabling T to be the current book and initially be the first book at the upper left corner of the image P;

s33, adding the book searching number of the T into the list A, and adding the coordinate of the T into the list POS;

s34, as shown in fig. 3, let T be the next book on the right side of the current book, and if there is no book on the right side, let T be the leftmost book on the next layer in the image P; if the current book is the last book in the bottom right corner of the image P, the process ends, otherwise, the process goes to step S33.

S4, scanning and analyzing the book searching numbers stored in the list A, and detecting the displaced book searching numbers, as shown in FIG. 4, the specific steps are as follows:

s41, creating an array L with the same length as the list a, and configured to record a longest increment subsequence length of each element in the list a with the element as an ending element, and assign a first element of the L array to 1, that is, L [1] ═ 1;

s42, creating an array P with the same length as the list A, recording the precursor elements of the elements in the list A in the longest increasing subsequence taking the elements as tail elements, and enabling P [1] to be NULL;

s43, defining a current subscript x, and making x equal to 2;

s44, detecting the index number of all elements before subscript x in list A, finding the longest increasing subsequence taking A [ x ] as the ending element, namely, for each i < x, calculating to obtain L [ x ] according to the following formula:

L[x]＝max{L(i)+1|A[i]<A[x]∧i<x} (2)

and recording P [ x ] as the previous element subscript i of the longest increment subsequence obtained when calculating L [ x ];

s45, making x equal to x +1, if x is less than or equal to | a |, turning to S44, otherwise, turning to S46;

s46, traversing the array L and finding the subscript k at the maximum value, and then tracing back the longest increasing subsequence taking A [ k ] as the tail element at P [ k ], and recording the longest increasing subsequence as LIS;

and S47, setting the misplaced book list B as all books of the list A except the LIS.

S5, marking the misplaced book label detected in the step S4 with a specific color according to the POS coordinate obtained in the step S3 in the image P.

And S6, repeating the steps from S1 to S5 until all bookshelves are processed.

By the method, all the misplaced books can be accurately found out, the image group marked with the misplaced books is generated, the image group can be stored in a computer or printed for viewing, and staff in a library can find corresponding bookshelves and books in the library according to the misplaced books marked in the image group to carry out arrangement work of the misplaced books. The system and the method for searching for the misplaced books have the advantages that the cost is low, the system and the method are not limited by the design of the library, the misplaced books can be found only by analyzing the image information of the bookshelf, the early preparation is simple, excessive maintenance is not needed in the follow-up process, and the system and the method are suitable for various large libraries.

The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.

Claims (3)

1. A method for searching a misplaced book is characterized by comprising the following steps:

s1, shooting a plurality of book images on the front side of the bookshelf, uploading the images to the system, and splicing the images into a complete image P in sequence;

s2, performing text recognition on the image P, and extracting book searching numbers on all spine labels in the image P;

s3, sequentially storing book searching numbers and coordinates of corresponding book labels in the image P into a list A and a list POS;

s4, scanning and analyzing the book searching numbers stored in the list A, and detecting the displaced book searching numbers;

s5, finding out coordinate data corresponding to the book searching number of the misplaced book detected in the step S4 in the list POS, and marking the image P with a specific color according to the coordinate data;

and S6, repeating the steps S1 to S5 until the staggered book searching and marking of all the bookshelves are completed.

2. The method for searching displaced books according to claim 1, wherein step S3 stores the book index number and the coordinates of the corresponding book label in the image P into the list a and the list POS in sequence, and is implemented by the following steps:

s31, constructing a list A for sequentially storing book searching numbers of the books and constructing a list POS for storing horizontal coordinates and vertical coordinates of book labels in the image P;

s32, enabling T to be the current book and initially be the first book at the upper left corner of the image P;

s33, adding the book searching number of the T into the list A, and adding the coordinate of the T into the list POS;

s34, let T be the next book on the right side of the current book, if there is no book on the right side, let T be the leftmost book on the next layer in the image P, if the current book is the last book on the lower right corner in the image P, then the process is ended, otherwise go to step S33.

3. The method for searching displaced books according to claim 1, wherein step S4 scans and analyzes the book search numbers stored in list a, and detects displaced book search numbers therein, which is specifically realized by the following steps:

s41, creating an array L with the same length as the list A, recording the longest increment subsequence length of each element in the list A taking the element as a tail element, and making L [1] equal to 1;

s42, creating an array P with the same length as the list A, recording the precursor elements of the elements in the list A in the longest increasing subsequence taking the elements as tail elements, and enabling P [1] to be NULL;

s43, defining a current subscript x, and making x equal to 2;

s44, detecting the index number of all elements before subscript x in list A, finding the longest increasing subsequence taking A [ x ] as the ending element, namely, for each i < x, calculating to obtain Lx according to the following formula:

L[x]＝maxLi+1|Ai＜Ax∧i＜x

and recording P [ x ] as the previous element subscript i of the longest increment subsequence obtained when calculating L [ x ];

s45, making x equal to x +1, if x is less than or equal to | a |, turning to S44, otherwise, turning to S46;

s46, traversing the array L and finding the subscript k at the maximum value, and then tracing back the longest increasing subsequence taking A [ k ] as the tail element at P [ k ], and recording the longest increasing subsequence as LIS;

and S47, setting the misplaced book list B as all books of the list A except the LIS.

Images