Book information processing method, device and equipment

Abstract

The present invention provides a method, device, and apparatus for processing book information. The method comprises: obtaining barcode data and cover data of a paper book; obtaining first characteristic data of the paper book based on the barcode data; matching the first characteristic data of the paper book with multiple pre-stored codes to obtain a matching result; and obtaining a target e-book corresponding to the paper book based on the matching result and the cover data. The present invention first obtains the barcode data and cover data of the paper book, then obtains the first characteristic data of the paper book using the obtained barcode data, then matches the first characteristic data of the paper book with multiple pre-stored codes, and obtains the target e-book corresponding to the paper book based on the matching result and the cover data, thereby improving the accuracy of searching for e-book resources corresponding to paper books.

Description

Technical Field

[0001] The present invention relates to the technical field of computer information processing, and also to a book information processing method, device and equipment. Background Art

[0002] With the development of electronic products, the rate of digital reading is increasing year by year. When reading the e-book corresponding to a paper book, it is necessary to quickly check whether the paper book has e-book resources and accurately and quickly find the corresponding e-book resources to improve the user experience. However, existing technologies have problems such as mismatch between paper books and the e-book resources found, and low accuracy. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a book information processing method, device and equipment to improve the accuracy of searching for e-book resources.

[0004] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0005] A first aspect of the present invention provides a method for processing book information, comprising:

[0006] Obtain the barcode data and cover data of the paper book;

[0007] Obtaining first characteristic data of the paper book according to the barcode data;

[0008] Matching the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result;

[0009] A target electronic book corresponding to the paper book is obtained according to the matching result and the cover data.

[0010] Optionally, obtaining first characteristic data of the paper book according to the barcode data includes:

[0011] Preprocessing the barcode data to obtain preprocessed barcode data;

[0012] Correcting the pre-processed barcode data to obtain standard barcode data;

[0013] The standard barcode data is decoded to obtain first characteristic data of the paper book.

[0014] Optionally, matching the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result includes:

[0015] The first characteristic data of the paper book is matched according to a plurality of pre-stored codes. If there is a pre-stored code identical to the first characteristic data of the paper book, the matching result is the same pre-stored code; otherwise, the matching result is empty.

[0016] Optionally, obtaining a target e-book corresponding to the paper book according to the matching result and the cover data includes:

[0017] When the matching result is the same pre-stored code, obtaining a target electronic book corresponding to the paper book according to the same pre-stored code;

[0018] When the matching result is empty, a target electronic book corresponding to the paper book is obtained according to the cover data.

[0019] Optionally, obtaining a target electronic book corresponding to the paper book according to the same pre-stored code includes:

[0020] Retrieving corresponding e-book data from a preset e-book database according to the same pre-stored code;

[0021] Verifying the e-book data to obtain a verification result;

[0022] When the verification result meets a preset qualification condition, the electronic book data is determined as a target electronic book corresponding to the paper book.

[0023] Optionally, obtaining a target e-book corresponding to the paper book according to the cover data includes:

[0024] Identify the cover data to obtain second characteristic data of the paper book;

[0025] determining keywords based on the second characteristic data of the paper book;

[0026] Retrieving corresponding e-book data from a preset e-book database according to the keyword;

[0027] Verifying the e-book data to obtain a verification result;

[0028] When the verification result meets a preset qualification condition, the electronic book data is determined as a target electronic book corresponding to the paper book.

[0029] Optionally, identifying the cover data to obtain second characteristic data of the paper book includes:

[0030] Adjusting the cover data to obtain adjusted cover data;

[0031] Performing feature extraction on the adjusted cover data according to a preset extraction method to obtain feature information;

[0032] The characteristic information is identified according to a preset identification method to obtain second characteristic data of the paper book.

[0033] A second aspect of the present invention provides a book information processing device, comprising:

[0034] An acquisition module, used to acquire barcode data and cover data of a paper book;

[0035] The processing module is configured to obtain first characteristic data of the paper book according to the barcode data of the paper book; match the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result; and obtain a target electronic book according to the matching result and the cover data of the paper book.

[0036] According to a third aspect of the present invention, a computing device is provided, comprising: a processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method according to the first aspect is executed.

[0037] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions, which, when executed on a computer, causes the computer to execute the method described in the first aspect.

[0038] The above solution of the present invention includes at least the following beneficial effects:

[0039] The above-mentioned solution of the present invention first obtains the barcode data and cover data of the paper book, then obtains the first characteristic data of the paper book through the obtained barcode data of the paper book, and then matches the first characteristic data of the paper book with multiple pre-stored codes. According to the matching results and the cover data, the target e-book corresponding to the paper book is obtained, thereby improving the accuracy of searching for e-book resources corresponding to the paper book. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 1 is a flow chart of a method for processing book information in an embodiment of the present invention;

[0041] Figure 2 It is a structural diagram of a book information processing device in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0043] like Figure 1 As shown, an embodiment of the present invention provides a book information processing method, comprising the following steps:

[0044] Step 101: Obtain barcode data and cover data of a paper book;

[0045] Step 102: Obtain first characteristic data of the paper book according to the barcode data;

[0046] Step 103: matching the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result;

[0047] Step 104: Obtain a target e-book corresponding to the paper book based on the matching result and the cover data.

[0048] The book information processing method proposed in an embodiment of the present invention first obtains the barcode data and cover data of the paper book, then obtains the first feature data of the paper book through the obtained barcode data of the paper book, and then matches the first feature data of the paper book with multiple pre-stored codes. Based on the matching results and the cover data, the target e-book corresponding to the paper book is obtained, thereby improving the accuracy of searching for e-book resources corresponding to the paper book.

[0049] In an optional embodiment of the present invention, the barcode data and cover data of the paper book obtained can be images, such as images of the barcode and cover of the paper book captured by an image capture device (such as a mobile phone or camera). It should be noted that although barcode data is used as an example in this embodiment, it should be understood that it is not limited to barcodes and can also be a QR code or other code that can identify the book number of the paper book (such as the International Standard Book Number (ISBN) code).

[0050] In an optional embodiment of the present invention, step 102 includes:

[0051] Step 1021, preprocessing the barcode data to obtain preprocessed barcode data;

[0052] Specifically, if the barcode data is image data, the image data needs to be preprocessed by filtering, smoothing, etc., wherein the image data can be filtered first to eliminate noise, reduce image details and improve image quality; then the image data can be smoothed by using a smoother or convolution kernel to eliminate noise and jagged effects (the edges of the image are not smooth and appear jagged).

[0053] Step 1022: Correct the pre-processed barcode data to obtain standard barcode data.

[0054] Specifically, the barcode in the image can be located first to obtain the barcode's position in the image and its dimensions, such as length and width. Then, based on the barcode's tilt angle, an image rotation algorithm can be used to rotate the barcode to the correct orientation so that it is aligned horizontally or vertically. The specific method is as follows: use a contour detection algorithm to find the outline of the barcode in the image, and from the detected contours, find the largest contour representing the barcode; use the minimum area bounding rectangle algorithm (or a similar algorithm) to find the barcode's bounding rectangle; calculate the tilt angle of the bounding rectangle, that is, the angle between the barcode and the horizontal direction; based on the calculated tilt angle (that is, the angle between the barcode and the horizontal direction), use an image rotation algorithm to rotate the barcode to the correct orientation so that it is aligned horizontally or vertically. If the barcode image is distorted (such as perspective distortion), image processing techniques such as affine transformation and perspective transformation can be used to correct the image distortion. Correcting the preprocessed barcode data can improve processing efficiency and the accuracy of subsequent interpretation results.

[0055] Step 1023: Decode the standard barcode data to obtain first characteristic data of the paper book.

[0056] Specifically, you can use existing open source libraries or commercial libraries that support barcode decoding, such as ZXing (barcode image processing library), LEADTOOLSSDK (image processing development toolkit), etc., input the standard barcode data into one of the decoding libraries, call the decoding function of the decoding library, decode the standard barcode data, and extract the book number (ISBN code) from the decoding result as the first feature data of the paper book.

[0057] In an optional embodiment of the present invention, step 103 includes:

[0058] The first characteristic data of the paper book is matched according to a plurality of pre-stored codes. If there is a pre-stored code identical to the first characteristic data of the paper book, the matching result is the same pre-stored code; otherwise, the matching result is empty.

[0059] Specifically, a preset e-book database (which can be a database, a platform, a website, etc., as long as it can realize the function of storing e-book resources) stores a large number of e-book resources and the codes corresponding to the e-books (pre-stored codes). Generally, one e-book uniquely corresponds to one pre-stored code. Therefore, you can search in the preset e-book database based on the first characteristic data of the paper book (i.e., the ISBN code). If you can find a pre-stored code that is consistent with it, it means that there may be an e-book corresponding to the paper book in the preset e-book database. Therefore, the matching result at this time is the pre-stored code that is consistent with the first characteristic data of the paper book. If you cannot find a pre-stored code that is consistent with the first characteristic data of the paper book in the preset e-book database, it means that there is no e-book corresponding to the paper book in the preset e-book database. At this time, the matching result is empty.

[0060] In an optional embodiment of the present invention, step 104 includes:

[0061] Step 1041: When the matching result is the same pre-stored code, a target electronic book corresponding to the paper book is obtained according to the same pre-stored code;

[0062] Specifically, the corresponding e-book may be retrieved from a preset e-book database according to a pre-stored code consistent with the first characteristic data of the paper book, and provided to the user as a target e-book corresponding to the paper book.

[0063] Step 1042: When the matching result is empty, obtain a target electronic book corresponding to the paper book according to the cover data.

[0064] Specifically, if the matching result is empty, it means that the e-book corresponding to the paper book may not exist in the preset e-book database, or the pre-stored code of the corresponding e-book in the preset e-book database may be inconsistent with the code of the paper book. Therefore, it is necessary to further search and confirm in the preset e-book database based on the cover data.

[0065] In an optional embodiment of the present invention, step 1041 includes:

[0066] Step 10411, retrieve corresponding e-book data from a preset e-book database according to the same pre-stored code;

[0067] Specifically, the corresponding e-book and its related information, such as author, publisher, publication date, etc., are retrieved from the preset e-book database according to the same pre-stored code. The e-book data includes the e-book and the author, publisher, publication date, etc. information.

[0068] Step 10412: verify the e-book data to obtain a verification result;

[0069] Specifically, the author, publisher, and publication date information in the e-book data can be compared with the author, publisher, and publication date information in the cover data. If all the information is consistent with the cover data, the verification result may include: consistent author, consistent publisher, and consistent publication date. If any one of the information is inconsistent, the verification result may include inconsistent information, such as at least one of the following: inconsistent author, inconsistent publisher, and inconsistent publication date.

[0070] It should be noted that the method for verifying the e-book data is not limited to the above method. You can also extract a certain section of content from the e-book data and compare it with the content extracted from the paper book. If they are consistent, the verification result is that the content is consistent. This application is only for example.

[0071] Step 10413: When the verification result meets the preset qualification condition, the electronic book data is determined as the target electronic book corresponding to the paper book.

[0072] Specifically, the preset qualification conditions may include: consistent author, consistent publisher, consistent publication date, etc. When the verification result matches the preset qualification conditions, it means that the obtained e-book is the electronic resource corresponding to the paper book. At this time, the e-book can be confirmed as the target e-book corresponding to the paper book and provided to the user. If the verification result does not match the preset qualification conditions, that is, the verification result is at least one of the following: the author of the e-book does not match the author of the paper book, the publisher of the e-book does not match the publisher of the paper book, or the publisher of the e-book does not match the publisher of the paper book, it means that the e-book is not the e-book corresponding to the paper book. At this time, a preset message (such as "No e-book corresponding to the paper book was found") can be provided to the user to inform the user of the search results.

[0073] In an optional embodiment of the present invention, step 1042 includes:

[0074] Step 10421: Identify the cover data to obtain second characteristic data of the paper book;

[0075] Specifically, the author, publisher, publication date and other information of the paper book can be identified from the cover data, and this information is used as the second feature data of the paper book for subsequent search of the corresponding e-book.

[0076] Step 10422: determining keywords based on the second characteristic data of the paper book;

[0077] Specifically, the second characteristic data of the paper book can be extracted according to the preset keywords, preset types, or preset formats, and words or numbers consistent with the preset keywords can be extracted as keywords. If the type of the preset keyword is digital, the digital data in the second characteristic data of the paper book is extracted as the keyword; if the preset keyword is "Author: XXX", "Author: A" in the second characteristic data of the paper book is extracted as the keyword. Here, there can be multiple preset keywords, preset types, or preset formats, and there can also be multiple determined keywords, which facilitates the subsequent accurate retrieval of corresponding e-book data from the preset e-book database and improves the accuracy rate.

[0078] Step 10423, retrieve corresponding e-book data from a preset e-book database according to the keyword;

[0079] Specifically, e-book data consistent with the keywords are retrieved from the preset e-book database. For example, if the keywords are "April 5, 2020", "Author: A", "Publisher: B Publisher", etc., the e-book data that matches "April 5, 2020", "Author: A", "Publisher: B Publisher" in the preset e-book database will be retrieved for subsequent verification.

[0080] Step 10424: verify the e-book data to obtain a verification result;

[0081] Specifically, the author, publisher, and publication date information in the e-book data can be compared with the author, publisher, and publication date information in the cover data. If all the information is consistent with the cover data, the verification result may include: consistent author, consistent publisher, and consistent publication date. If any one of the information is inconsistent, the verification result may include inconsistent information, such as at least one of the following: inconsistent author, inconsistent publisher, and inconsistent publication date.

[0082] It should be noted that the method for verifying the e-book data is not limited to the above method. You can also extract a certain section of content from the e-book data and compare it with the content extracted from the paper book. If they are consistent, the verification result is that the content is consistent. This application is only for example.

[0083] Step 10425: When the verification result meets the preset qualification condition, the electronic book data is determined as the target electronic book corresponding to the paper book.

[0084] Specifically, the preset qualification conditions may include: consistent author, consistent publisher, consistent publication date, etc. When the verification result matches the preset qualification conditions, it means that the obtained e-book is the electronic resource corresponding to the paper book. At this time, the e-book can be confirmed as the target e-book corresponding to the paper book and provided to the user. If the verification result does not match the preset qualification conditions, that is, the verification result is at least one of the following: the author of the e-book does not match the author of the paper book, the publisher of the e-book does not match the publisher of the paper book, or the publisher of the e-book does not match the publisher of the paper book, it means that the e-book is not the e-book corresponding to the paper book. At this time, a preset message (such as "No e-book corresponding to the paper book was found") can be provided to the user to inform the user of the search results.

[0085] In an optional embodiment of the present invention, step 10421 includes:

[0086] Step 104211, adjusting the cover data to obtain adjusted cover data;

[0087] Specifically, the cover data may be a collected cover image of a paper book, and the cover image may be adjusted in at least one of the following ways:

[0088] By enhancing the contrast and brightness of the cover image, the text in the cover image is made clearer, which facilitates subsequent feature extraction and recognition; the cover image is converted into a black and white binary image, that is, the pixels in the cover image have only two types, black and white, which helps to simplify the subsequent processing process; noise in the cover image, such as spots, scratches, etc., is removed to improve the quality of the cover image; the tilted cover image is corrected to keep the text level and avoid errors in subsequent recognition.

[0089] Step 104212: performing feature extraction on the adjusted cover data according to a preset extraction method to obtain feature information;

[0090] Specifically, feature extraction is performed on the adjusted cover data to obtain relevant information about the paper book, such as the author, publisher, and publication date, for subsequent recognition. Here, the feature information can be an image, such as an image with the author, publisher, or publication date. The type and format of the feature information can be selected based on the subsequent recognition method used.

[0091] The default extraction method can be any of the following:

[0092] Structure-based feature extraction: By analyzing the structural features of text / numbers, such as strokes and radicals, the features of text / numbers in the cover data are extracted. This method is suitable for identifying text / numbers with clear and standardized structures.

[0093] Statistical feature extraction: The features of text / numbers are extracted by counting the pixel distribution, grayscale value and other features of the text / numbers. This method is suitable for identifying complex and changeable text / numbers.

[0094] Step 104213: Identify the characteristic information according to a preset identification method to obtain second characteristic data of the paper book.

[0095] Specifically, identifying the characteristic information mainly involves identifying the text or numbers in the characteristic information as the second characteristic data of the paper book, so as to facilitate the subsequent rapid and accurate retrieval of corresponding e-book data from a preset e-book database based on the second characteristic data of the paper book.

[0096] The preset identification method can be any of the following:

[0097] (1) Template matching preset recognition method

[0098] This method is to compare the extracted text features (i.e., feature information) with pre-stored character templates one by one to find the most similar character template, thereby recognizing the text / number. This recognition method is simple and intuitive.

[0099] (2) Using the preset recognition method of the preset recognition model

[0100] First, multiple character or digital features extracted from the image are collected as input, and the corresponding character or digital labels are used as output to obtain a training data set, which is divided into a training set and a validation set. Then, the initial parameters of the preset network model are set, such as the initial parameters including: an initial learning rate of 0.001 or 0.01, a batch size (the number of training data processed by the preset network model each time) of 32 or 64 or 128 or 256, and the number of iterations of 50 or 100. The various parameters are set specifically according to the preset network model used, such as the initial parameters also including weights and bias items. The training set is then used to train the preset network model with the various parameters set, and the various parameters of the preset network model are adjusted so that the output of the preset network model approaches the true label.

[0101] The method for adjusting the parameters of the preset network model is as follows:

[0102] Input the training set into the input layer of the preset network model; weight and sum the output of the input layer through the activation function to generate the output of at least one processing layer; weight and sum the output of the last processing layer again, and generate the final output through the activation function; use the loss function Calculate the error between the predicted value output by the preset network model and the true label; calculate the partial derivative of the loss function with respect to the final output to obtain the gradient of the output layer; use the chain rule to propagate the gradient of the output layer backward to at least one processing layer and calculate the gradient of each processing layer. This step is repeated in each processing layer of the preset network model until the input layer is reached; based on the gradient of each processing layer, calculate the gradient of all parameters in the preset network model, including weights and bias terms. Using the formula: The adjusted parameters are obtained. The learning rate is a preset hyperparameter that controls the step size of each parameter update. Based on the changes in the loss function and the performance on the validation set, parameters such as the learning rate and batch size are adjusted, and appropriate stopping conditions (such as the number of iterations and loss threshold) are set to terminate the training process. The preset recognition model is obtained based on the final adjusted parameters.

[0103] Among them, L is the loss function, N is the total number of samples in the training set, C is the total number of corresponding character or digital labels, and y i,c is the value of the cth category in the true label of the i-th sample (0 or 1), is the value of the cth category in the predicted value of the i-th sample. t+1 is the parameter at the t+1th iteration, θ t is the parameter at the tth iteration, η is the learning rate, is the gradient of the loss function with respect to the parameters, x i is a randomly selected sample, y i are the labels of randomly selected samples.

[0104] The preset recognition model obtains the second feature data of the paper book according to the input feature information. The specific process includes, that is, step 104213 includes the following steps:

[0105] Step 1042131: input the feature information into the input layer of the preset recognition model to obtain a first output result;

[0106] Here, the input layer mainly converts the feature information into a format or type that can be recognized and processed by the preset recognition model. If the feature information is an image with an author, publisher, or publication date, the input layer can be The feature information is processed to obtain a first output result, where N is the first output result, W is the feature information, F is the size of the filter, P is the padding size, that is, the number of layers of zeros padded on the boundary of the feature information, which is used to control the size of the first output result, and S is the step size, that is, the distance the filter slides on the input feature information.

[0107] To illustrate with a specific example, assume the feature information is an image with the author, with a length and width of 227, three color channels (RGB), a filter with a length and width of 11, a stride of 4, and a padding of 0. Then, according to the above formula, the size of the first output result can be calculated as: N = (227 - 11 + 2 × 0) / 4 + 1 = 55, that is, the length and width of the first output result are both 55.

[0108] Step 1042132: input the first output result into at least one processing layer of a preset recognition model to obtain a second output result;

[0109] It should be noted that the preset recognition model may include multiple processing layers, and the first output result is sequentially input into the multiple processing layers for processing. The processing process of the processing layer on the first output result includes:

[0110] Use multiple filters to perform local connections and feature extraction on the first output result. Each filter slides over the first output result, calculating the dot product between the filter and the local area of ​​the first output result to generate a feature map.

[0111] Downsample the feature map to reduce the dimension and computational complexity of the data while retaining important feature information as features to be processed;

[0112] All features to be processed are combined to obtain a fixed-size vector as the second output result. The second output result includes all important features extracted from the feature information.

[0113] Step 1042133: input the second output result into the output layer of the preset recognition model to obtain the second feature data of the paper book.

[0114] Specifically, by calculating the dot product of the input second output result and the weight vector of each neuron in the output layer, and adding the bias term, we get the score or probability corresponding to the category, and then use the softmax function The scores are converted into probability distributions. Therefore, for character or number recognition, the prediction result of the preset recognition model is usually a probability distribution, which indicates the probability of each character or number appearing.

[0115] Among them, y i is the value of the i-th element in the second output after the softmax function, that is, the probability prediction of the i-th category, x i is the i-th element in the second output result, is the index value of the i-th element in the second output result, is the sum of the index values ​​of all elements in the second output result.

[0116] A specific embodiment of the book information processing method proposed in the embodiment of the present invention is:

[0117] Step 111: Acquire paper book data.

[0118] The barcode image and the cover image of the paper book are captured by an image capture device (such as a mobile phone, a camera, etc.) as the barcode data and the cover data of the paper book respectively.

[0119] Step 112: first characteristic data of the paper book.

[0120] The barcode image is pre-processed by filtering, smoothing, etc., and then corrected. After that, it is decoded to obtain the ISBN code of the paper book, that is, the first characteristic data of the paper book.

[0121] Step 113: Match the first characteristic data of the paper book with a plurality of pre-stored codes.

[0122] Search the e-book resources stored in the preset e-book database for a pre-stored code that is consistent with the ISBN code of the paper book. If found, the matching result is the same pre-stored code; otherwise, the matching result is empty.

[0123] Here, the preset e-book database is a database containing e-book resource information, including cover information of each e-book (such as author, publisher and publication date, etc.). For example, the cover size of each e-book is set to 300x400 pixels for easy storage.

[0124] Step 114: Determine the target e-book.

[0125] First, if the matching result is the same pre-stored code, the e-book corresponding to the pre-stored code is directly retrieved from the preset e-book database. If the matching result is empty, the author, publisher, publication date and other information of the paper book are identified from the cover data as the second characteristic data of the paper book, and the e-book matching the second characteristic data of the paper book is searched in the preset e-book database. After the e-book is found, it needs to be verified based on the author, publisher, publication date, chapter / paragraph content and other information. If the verification is passed, the found e-book is confirmed to be the target e-book.

[0126] Here, if no e-book matching the second characteristic data of the paper book can be found in the preset e-book database, the e-book with the highest similarity can be selected as the target e-book based on the similarity between the e-book cover and the second characteristic data of the paper book in the preset e-book database.

[0127] Afterwards, the target e-book may be displayed in a preset display mode (such as displaying only the cover of the e-book), or the target e-book may be sent to the client for the user to view.

[0128] The book information processing method proposed in the embodiment of the present invention combines cover character recognition and image similarity matching to improve the accuracy of book recognition. It also supports book cover recognition and back ISBN barcode recognition to adapt to different situations.

[0129] The paper book identifier contains a barcode, which contains the ISBN code of the paper book. The barcode corresponds to the paper book and is unique. The e-book corresponding to the paper book is found through the ISBN code. Since the same book title corresponds to different paper books and has different ISBN numbers, when the corresponding e-book cannot be found through the ISBN number, the information contained in the book cover can be extracted, including text information and image element information. The valid text content is identified based on the extracted information, and the e-book is matched in the preset e-book database.

[0130] The book information processing method proposed in the embodiment of the present invention obtains the barcode data and cover data of the paper book by scanning the cover and barcode of the paper book, and preferentially matches the e-book according to the ISBN number and book cover information in the barcode data, so that the paper book is associated with the online e-book, thereby improving the accuracy of searching for the e-book corresponding to the paper book and solving the problem of mismatch between paper book and e-book resources.

[0131] like Figure 2 As shown, an embodiment of the present invention provides a book information processing device 200, comprising:

[0132] An acquisition module 201 is used to acquire barcode data and cover data of a paper book;

[0133] The processing module 202 is configured to obtain first characteristic data of the paper book based on the barcode data of the paper book; match the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result; and obtain a target electronic book based on the matching result and the paper book cover data.

[0134] Optionally, obtaining first characteristic data of the paper book according to the barcode data includes:

[0135] Preprocessing the barcode data to obtain preprocessed barcode data;

[0136] Correcting the pre-processed barcode data to obtain standard barcode data;

[0137] The standard barcode data is decoded to obtain first characteristic data of the paper book.

[0138] Optionally, matching the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result includes:

[0139] The first characteristic data of the paper book is matched according to a plurality of pre-stored codes. If there is a pre-stored code identical to the first characteristic data of the paper book, the matching result is the same pre-stored code; otherwise, the matching result is empty.

[0140] Optionally, obtaining a target e-book corresponding to the paper book according to the matching result and the cover data includes:

[0141] When the matching result is the same pre-stored code, obtaining a target electronic book corresponding to the paper book according to the same pre-stored code;

[0142] When the matching result is empty, a target electronic book corresponding to the paper book is obtained according to the cover data.

[0143] Optionally, obtaining a target electronic book corresponding to the paper book according to the same pre-stored code includes:

[0144] Retrieving corresponding e-book data from a preset e-book database according to the same pre-stored code;

[0145] Verifying the e-book data to obtain a verification result;

[0146] When the verification result meets a preset qualification condition, the electronic book data is determined as a target electronic book corresponding to the paper book.

[0147] Optionally, obtaining a target e-book corresponding to the paper book according to the cover data includes:

[0148] Identify the cover data to obtain second characteristic data of the paper book;

[0149] determining keywords based on the second characteristic data of the paper book;

[0150] Retrieving corresponding e-book data from a preset e-book database according to the keyword;

[0151] Verifying the e-book data to obtain a verification result;

[0152] When the verification result meets a preset qualification condition, the electronic book data is determined as a target electronic book corresponding to the paper book.

[0153] Optionally, identifying the cover data to obtain second characteristic data of the paper book includes:

[0154] Adjusting the cover data to obtain adjusted cover data;

[0155] Performing feature extraction on the adjusted cover data according to a preset extraction method to obtain feature information;

[0156] The characteristic information is identified according to a preset identification method to obtain second characteristic data of the paper book.

[0157] The book information processing device proposed in an embodiment of the present invention first obtains the barcode data and cover data of a paper book, then obtains the first feature data of the paper book through the obtained barcode data of the paper book, and then matches the first feature data of the paper book with multiple pre-stored codes. Based on the matching results and the cover data, a target e-book corresponding to the paper book is obtained, thereby improving the accuracy of searching for e-book resources corresponding to the paper book.

[0158] It should be noted that the device is a device corresponding to the above method, and all implementations in the above method embodiment are applicable to the embodiment of the device and can achieve the same technical effects, which will not be described in detail in this embodiment.

[0159] An embodiment of the present invention further provides a computing device comprising: a processor and a memory storing a computer program. When the computer program is executed by the processor, the computer program performs the method described in any of the above embodiments. All implementations in the above method embodiments are applicable to the embodiments of this device and can achieve the same technical effects. These are not further described in this embodiment.

[0160] An embodiment of the present invention further provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described in any of the above embodiments. All implementations in the above method embodiments are applicable to the embodiments of the device and can achieve the same technical effects. These are not further described in this embodiment.

[0161] It should be noted that in the above embodiments, the descriptions of the methods and steps, the order of the steps, the sequence numbers of the steps, etc. according to the embodiments of the present invention are merely exemplary and are not intended to limit the present invention. The terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or device comprising that element. Furthermore, it should be noted that the scope of the methods and devices in the implementations of the above embodiments is not limited to performing functions in the order shown or discussed, but may also include performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.

[0162] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A book information processing method, characterized in that: include: Obtain the barcode data and cover data of the paper book; Obtaining first characteristic data of the paper book according to the barcode data; Matching the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result; Obtaining a target e-book corresponding to the paper book according to the matching result and the cover data; The first characteristic data of the paper book is matched with a plurality of pre-stored codes to obtain a matching result, including: Matching the first characteristic data of the paper book according to a plurality of pre-stored codes, if there is a pre-stored code identical to the first characteristic data of the paper book, the matching result is the same pre-stored code; otherwise, the matching result is empty; Wherein, obtaining a target e-book corresponding to the paper book according to the matching result and the cover data includes: When the matching result is the same pre-stored code, obtaining a target electronic book corresponding to the paper book according to the same pre-stored code; When the matching result is empty, obtaining a target e-book corresponding to the paper book according to the cover data; Wherein, obtaining a target e-book corresponding to the paper book according to the cover data includes: Identify the cover data to obtain second characteristic data of the paper book; determining keywords based on the second characteristic data of the paper book; Retrieving corresponding e-book data from a preset e-book database according to the keyword; Verifying the e-book data to obtain a verification result; When the verification result meets a preset qualification condition, determining the electronic book data as a target electronic book corresponding to the paper book; The cover data is identified to obtain the second characteristic data of the paper book, including: Adjusting the cover data to obtain adjusted cover data; Performing feature extraction on the adjusted cover data according to a preset extraction method to obtain feature information; Identify the characteristic information according to a preset identification method to obtain second characteristic data of the paper book; The feature information is identified according to a preset identification method to obtain the second feature data of the paper book, including: The feature information is input into the input layer of the preset recognition model to obtain the first output result; the input layer is Processing the feature information to obtain a first output result, where N is the first output result, W is the feature information, F is the length or width of the filter, P is the padding size, and S is the step size; Inputting the first output result into at least one processing layer of a preset recognition model to obtain a second output result; performing local connection and feature extraction on the first output result using multiple filters to generate a feature map; downsampling the feature map to obtain features to be processed; integrating all the features to be processed together to obtain a vector of fixed size as the second output result; The second output result is input into the output layer of the preset recognition model to obtain the second feature data of the paper book.

2. The book information processing method according to claim 1, characterized in that According to the barcode data, first characteristic data of the paper book is obtained, including: Preprocessing the barcode data to obtain preprocessed barcode data; Correcting the pre-processed barcode data to obtain standard barcode data; The standard barcode data is decoded to obtain first characteristic data of the paper book.

3. The book information processing method according to claim 1, wherein: Obtaining a target electronic book corresponding to the paper book according to the same pre-stored code includes: Retrieving corresponding e-book data from a preset e-book database according to the same pre-stored code; Verifying the e-book data to obtain a verification result; When the verification result meets a preset qualification condition, the electronic book data is determined as a target electronic book corresponding to the paper book.

4. A book information processing device, characterized in that: include: An acquisition module, used to acquire barcode data and cover data of a paper book; a processing module, configured to obtain first characteristic data of the paper book according to the barcode data of the paper book; Matching the first characteristic data of the paper book with a plurality of pre-stored codes to obtain a matching result; obtaining a target electronic book based on the matching result and the paper book cover data; The first characteristic data of the paper book is matched with a plurality of pre-stored codes to obtain a matching result, including: Matching the first characteristic data of the paper book according to a plurality of pre-stored codes, if there is a pre-stored code identical to the first characteristic data of the paper book, the matching result is the same pre-stored code; otherwise, the matching result is empty; Wherein, obtaining a target e-book corresponding to the paper book according to the matching result and the cover data includes: When the matching result is the same pre-stored code, obtaining a target electronic book corresponding to the paper book according to the same pre-stored code; When the matching result is empty, obtaining a target e-book corresponding to the paper book according to the cover data; Wherein, obtaining a target e-book corresponding to the paper book according to the cover data includes: Identify the cover data to obtain second characteristic data of the paper book; determining keywords based on the second characteristic data of the paper book; Retrieving corresponding e-book data from a preset e-book database according to the keyword; Verifying the e-book data to obtain a verification result; When the verification result meets a preset qualification condition, determining the electronic book data as a target electronic book corresponding to the paper book; The cover data is identified to obtain the second characteristic data of the paper book, including: Adjusting the cover data to obtain adjusted cover data; Performing feature extraction on the adjusted cover data according to a preset extraction method to obtain feature information; Identify the characteristic information according to a preset identification method to obtain second characteristic data of the paper book; The feature information is identified according to a preset identification method to obtain the second feature data of the paper book, including: The feature information is input into the input layer of the preset recognition model to obtain the first output result; the input layer is Processing the feature information to obtain a first output result, where N is the first output result, W is the feature information, F is the length or width of the filter, P is the padding size, and S is the step size; Inputting the first output result into at least one processing layer of a preset recognition model to obtain a second output result; performing local connection and feature extraction on the first output result using multiple filters to generate a feature map; downsampling the feature map to obtain features to be processed; integrating all the features to be processed together to obtain a vector of fixed size as the second output result; The second output result is input into the output layer of the preset recognition model to obtain the second feature data of the paper book.

5. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method according to any one of claims 1 to 3 is performed.

6. A computer-readable storage medium, characterized in that The device stores instructions, which, when executed on a computer, enable the computer to execute the method according to any one of claims 1 to 3.

Images