CN111325195A - Text recognition method, device and electronic device - Google Patents

Abstract

The embodiment of the specification discloses a text recognition method, a text recognition device and electronic equipment. Extracting block features of the text blocks aiming at the text blocks, and judging whether the block features of two adjacent lines of text blocks reach a preset feature condition, wherein the preset feature condition is a feature condition which is established by using a training sample and is met by the block features of the two adjacent lines of text blocks when the two adjacent lines of text blocks belong to the same text information. And determining the operation on the two adjacent lines of text blocks according to the judgment result, wherein the operation comprises one of text information combination and non-combination.

Description

Text recognition method, device and electronic device

technical field

The embodiments of this specification relate to the field of computer technology, and in particular, to a text recognition method, apparatus, and electronic device.

Background technique

In daily life, it is often necessary to perform text recognition on carriers such as pictures, and then perform text processing according to the recognition results. Then, how to identify complete text information is a topic that has been discussed in the industry.

SUMMARY OF THE INVENTION

In view of this, the embodiments of this specification provide a text recognition method, apparatus and electronic device for accurately recognizing text information.

The embodiments of this specification adopt the following technical solutions:

The embodiments of this specification provide a text recognition method, including:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

It is judged whether the block features of the text blocks in two adjacent lines meet a preset feature condition, and the preset feature condition is that when two adjacent lines of text blocks belong to the same text information established by using training samples, the adjacent text blocks belong to the same text information. The feature conditions satisfied by the block feature of the two-line text block;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The embodiment of this specification also provides a text recognition method, including:

Perform text recognition on the training samples, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

The merging model is trained by using the block features of the text blocks to determine the preset feature conditions in the merging model, so that when the block features of two adjacent lines of text blocks in the object to be recognized are identified, the whether the block feature of the text block meets the preset feature condition, and perform an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and not merging.

The embodiment of this specification also provides a text recognition method, including:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

A merging model is used to process the block features of the text blocks in two adjacent lines to obtain a judgment result of whether the text blocks in the two adjacent lines meet the preset feature conditions. The block features of the two adjacent lines of text blocks obtained are obtained by training to determine the preset feature conditions;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The embodiments of this specification also provide a text recognition device, including:

A text recognition module, which performs text recognition on the object to be recognized, and obtains text blocks based on the recognized lines of text;

an extraction module to extract block features of the text block;

A judgment module, for judging whether the block features of the text blocks in two adjacent lines meet a preset feature condition, and the preset feature condition is established by using a training sample when the text blocks in two adjacent lines belong to the same text information, the The feature conditions satisfied by the block features of the two adjacent lines of text blocks;

The execution module performs an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The embodiments of this specification also provide a text recognition device, including:

The text recognition module performs text recognition on the training samples, and obtains text blocks based on the recognized lines of text;

an extraction module to extract block features of the text block;

The training module uses the block features of the text blocks to train the merged model to determine the preset feature conditions in the merged model, so that when identifying the block features of two adjacent lines of text blocks in the object to be identified, it is possible to judge the adjacent Whether the block feature of the two lines of the text block meets the preset feature condition, an operation is performed on the two adjacent lines of the text block according to the judgment result, and the operation includes one of merging and not merging.

The embodiments of this specification also provide a text recognition device, including:

A text recognition module, which performs text recognition on the object to be recognized, and obtains text blocks based on the recognized lines of text;

an extraction module to extract block features of the text block;

The model processing module uses a merged model to process the block features of the text blocks in two adjacent lines, and obtains a judgment result of whether the text blocks in the two adjacent lines meet the preset feature conditions. The block features of the two adjacent lines of text blocks identified in the training samples are trained to determine the preset feature conditions;

The execution module performs an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The embodiments of this specification also provide an electronic device, including:

processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

It is judged whether the block features of the text blocks in two adjacent lines meet a preset feature condition, and the preset feature condition is that when two adjacent lines of text blocks belong to the same text information established by using training samples, the adjacent text blocks belong to the same text information. The feature conditions satisfied by the block feature of the two-line text block;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The embodiments of this specification also provide an electronic device, including:

processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

Perform text recognition on the training samples, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

The merging model is trained by using the block features of the text blocks to determine the preset feature conditions in the merging model, so that when the block features of two adjacent lines of text blocks in the object to be recognized are identified, the whether the block feature of the text block meets the preset feature condition, and perform an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and not merging.

The embodiments of this specification also provide an electronic device, including:

processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

A merging model is used to process the block features of the text blocks in two adjacent lines to obtain a judgment result of whether the text blocks in the two adjacent lines meet the preset feature conditions. The block features of the two adjacent lines of text blocks obtained are obtained by training to determine the preset feature conditions;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The above-mentioned at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects:

The embodiments of this specification provide an automatic recognition solution for whether texts are merged. The essential technical solution is that, by performing text recognition on each line of text in an object to be recognized, text blocks can be obtained based on the recognized lines of text. Extract the block feature of the text block for the text block, and determine whether the block features of the two adjacent lines of text blocks meet a preset feature condition, where the preset feature condition is that the two adjacent lines of text blocks belong to In the case of the same text information, the feature condition satisfied by the block features of the text blocks of the two adjacent lines. An operation on the text blocks of the two adjacent lines is determined according to the judgment result, and the operation includes one of merging into text information and not merging.

Therefore, this scheme can automatically identify whether two adjacent lines of text blocks belong to the same text information by using the block features of the text blocks, combined with the preset feature conditions established in the training samples, and determine whether the adjacent two lines of text blocks belong to the same text information according to the judgment result. Line text blocks are merged. The solutions described in the embodiments of this specification can improve text recognition efficiency.

Description of drawings

The drawings described herein are used to provide further understanding of the embodiments of the present specification, and constitute a part of the embodiments of the present specification. The schematic embodiments and descriptions of the present specification are used to explain the present application and do not constitute an improper limitation of the present application. In the attached image:

FIG. 1 is a schematic diagram of a system architecture of a text recognition scheme proposed by an embodiment of this specification;

2 is a flowchart of a text recognition method provided by an embodiment of the present specification;

3 is a schematic diagram of a text block in a text recognition process provided by an embodiment of the present specification;

4 is a schematic diagram of a text block in a text recognition process provided by an embodiment of the present specification;

FIG. 5 is a schematic diagram of a text block in a text recognition process provided by an embodiment of the present specification;

6 is a flowchart of an application example of a text recognition method proposed in an embodiment of this specification;

7 is a flowchart of a text recognition method provided by an embodiment of the present specification;

8 is a flowchart of a text recognition method provided by an embodiment of the present specification;

9 is a flowchart of an application example of a text recognition method provided by an embodiment of the present specification;

10 is a flowchart of an application example of a text recognition method provided by an embodiment of this specification;

FIG. 11 is a schematic structural diagram of a text recognition device according to an embodiment of the present specification;

12 is a schematic structural diagram of a text recognition device according to an embodiment of the present specification;

FIG. 13 is a schematic structural diagram of a text recognition apparatus according to an embodiment of the present specification.

Detailed ways

It is found by analyzing the prior art that the prior art provides a single character recognition technology, for example, an optical character recognition technology is used for character recognition. Afterwards, the recognized characters can be merged to form complete text information in an artificial manner.

The embodiments of this specification propose a text recognition method, device, and electronic device. The essential technical solution is that, by performing text recognition on each line of text in an object to be recognized, text blocks can be obtained based on each line of recognized text. Extract the block feature of the text block for the text block, and determine whether the block features of the two adjacent lines of text blocks meet a preset feature condition, where the preset feature condition is that the two adjacent lines of text blocks belong to In the case of the same text information, the feature condition satisfied by the block features of the text blocks of the two adjacent lines. An operation on the text blocks of the two adjacent lines is determined according to the judgment result, and the operation includes one of merging into text information and not merging.

Therefore, the embodiments of this specification provide an automatic identification scheme for whether texts are merged. This scheme can automatically identify whether two adjacent lines of text blocks belong to a text block by using the block features of the text blocks, combined with the preset feature conditions established in the training samples. The same text information, and whether to merge the two adjacent lines of text blocks is determined according to the judgment result. The solutions described in the embodiments of this specification can improve text recognition efficiency.

In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the specific embodiments of the present specification and the corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present specification, but not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

The technical solutions provided by the embodiments of the present specification will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic diagram of a system architecture of a text recognition solution proposed in an embodiment of the present specification.

As shown in FIG. 1 , the system architecture 100 may include terminal devices 101 , 102 , and 103 , a network 104 and a server 105 . The network 104 is a medium used to provide a communication link between the terminal devices 101 , 102 , 103 and the server 105 . The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The terminal devices 101, 102, and 103 interact with the server 105 through the network 104 to receive or send messages and the like. Various client applications may be installed on the terminal devices 101 , 102 and 103 . For example, browser applications, search applications, instant messaging tools, etc.

The terminal devices 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, and 103 are hardware, they can be various electronic devices, including but not limited to smart phones, tablet computers, e-book readers, laptop computers, desktop computers, and the like. When the terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (eg, multiple software or software modules for providing distributed services), or as a single software or software module. There is no specific limitation here.

The server 105 may be a server that provides various services, such as a back-end server that tests client applications installed on the terminal devices 101 , 102 , and 103 . It should be noted that the text recognition method provided by the embodiments of the present disclosure is generally executed by the server 105 , and accordingly, the text recognition apparatus is generally provided in the server 105 . At this time, the terminal devices 101, 102, 103 and the network 104 may not exist.

It should also be noted that the testing of the client applications installed on the terminal devices 101 , 102 and 103 may also be performed by the terminal devices 101 , 102 and 103 . At this time, the text recognition method may be executed by the terminal devices 101 , 102 and 103 , and correspondingly, the text recognition apparatus may also be provided in the terminal devices 101 , 102 and 103 . At this point, server 105 and network 104 may be absent from example system architecture 100 .

It should be noted that the server 105 may be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or can be implemented as a single server. When the server 105 is software, it can be implemented as multiple software or software modules (for example, multiple software or software modules for providing distributed services), or can be implemented as a single software or software module. There is no specific limitation here.

It should be understood that the numbers of terminal devices, networks and servers in FIG. 1 are merely illustrative. There can be any number of terminal devices, networks and servers according to implementation needs.

FIG. 2 is a flowchart of a text recognition method proposed in an embodiment of the present specification.

Step 201: Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text.

In a specific application, text information can be displayed in the object to be recognized. The object to be recognized described in the embodiment of this specification may be an image, and the image may be obtained by scanning a carrier carrying text information, or may be obtained by digital synthesis, which is not specifically limited herein.

To perform text recognition on each line of text in the object to be recognized, the optical character recognition technology OCR (full name: Optical Character Recognition) may be used specifically. OCR refers to the process in which electronic devices (such as scanners or digital cameras) examine characters printed on paper, determine their shapes by detecting dark and light patterns, and then translate the shapes into computer text using character recognition methods. text in the image.

In the embodiment of this specification, the object to be recognized may include multiple lines of text. The lines here do not define a specific direction, but represent a group of words arranged in a straight direction. In the embodiment of this specification, character recognition is performed on each line of text in the object to be recognized, and text blocks are obtained based on the recognized lines of text, which may include:

Perform text recognition on the object to be recognized;

It also recognizes the text arrangement in the object to be recognized, and obtains the recognized lines of text;

Mark the recognized lines of text as text blocks.

The text block described in the embodiments of this specification is a text unit, and does not limit the text structure.

The embodiments of the present specification perform subsequent processing on the recognized characters in units of text blocks.

Step 203: Extract the block feature of the text block.

In the embodiment of this specification, the block feature is a feature possessed by the text block itself. The innovation of the text block automatic identification and merging scheme proposed in the embodiments of this specification is that the block feature of the text block is identified by taking the text block as a unit, and combined with the block features of the two adjacent lines of text blocks described later, A judgment result is made as to whether the text blocks belong to the same text information.

In the embodiment of this specification, the block feature may include one or both of the line height of each text block and the line spacing between two adjacent lines of the text blocks, which are not specifically limited herein.

In the embodiment of this specification, extracting the block feature of the text block may include:

Create a coordinate space;

Identify the coordinate value of the text block in the coordinate space;

A block feature of the text block is determined based on the coordinate values.

Specifically, the line height of the text block, the line spacing between two adjacent lines of text blocks, or other dimensions can be calculated by using the coordinate values, which are not specifically limited herein.

Step 205: Determine whether the block features of the text blocks in two adjacent lines meet a preset feature condition, where the preset feature condition is established by using training samples when two adjacent lines of text blocks belong to the same text information. The feature conditions that are satisfied by the block features of the two adjacent lines of text blocks.

The embodiment of the present specification uses the analysis or learning of the training samples to extract preset feature conditions. The preset feature conditions are that when two adjacent lines of text blocks belong to the same text information, the block features of the two adjacent lines of text blocks have The feature threshold condition of , the preset feature condition can be used as the text block merging condition. This provides the possibility to realize the automatic text block merging solution proposed by the embodiments of this specification.

The text information described in the embodiments of the present specification refers to paragraphs or documents composed of words that describe specified matters. The fact that two adjacent lines of text blocks belong to the same text information means that the texts contained in the two adjacent lines of text blocks describe the same matter, and can be combined into one paragraph or document formally.

Specifically, in another example of this specification, judging whether the block features of two adjacent lines of text blocks meet the preset feature conditions, including:

It is determined that the line height of the text blocks in two adjacent lines is not less than the line spacing.

In this case, the line height of two adjacent lines of text blocks is not less than the line spacing, which is an example of a preset feature condition.

The principle is that if the current actual line height is not less than the line spacing, that is, the line height is close to the line spacing, or the line height is greater than the line spacing, and two adjacent lines of text blocks are close to each other, then the two adjacent lines of text blocks have a high probability of belonging to the same text information. Conversely, if the current actual line height is smaller than the line spacing, that is, the line spacing is large, and the distance between two adjacent lines of text blocks is large, then the two adjacent lines of text blocks are likely not to belong to the same text information.

In this example, the line heights of two adjacent lines of text blocks are close or equal.

Wherein, judging that the line height of the text blocks in the two adjacent lines is not less than the line spacing, including:

It is judged whether the line height of the text blocks in the two adjacent lines exceeds the line spacing by a preset difference.

In this case, the preset difference value is one of preset characteristic conditions.

The principle is that if the line height exceeds the line spacing and reaches the preset difference, that is, the word height in the text block is much larger than the line spacing, and the words are large and the line spacing is small, then the two adjacent lines of text blocks have a high probability of belonging to the same text information. Conversely, if the line height exceeds the line spacing, or the line height exceeds the line spacing but does not exceed the preset difference, that is, the word height in the text block is smaller than the line spacing, and the words are small but the line spacing is large, then the two adjacent lines of text blocks are not likely to belong to the same text message.

In another example of this specification, judging whether the block features of the text blocks in two adjacent lines meet a preset feature condition includes:

If the block feature of the text block includes a line height, it is determined whether the difference between the line heights of the text blocks in two adjacent lines is not greater than a preset line height difference.

In this case, the preset line height difference is one of the preset characteristic conditions.

The principle is that if the difference between the line heights is not greater than the preset line height difference, that is, the line heights of two adjacent lines of text blocks are close to or equal, then the two adjacent lines of text blocks have a high probability of belonging to the same text information. If the difference between the line heights is greater than the preset height difference, that is, the difference between the line heights is large, there is a high probability that the two adjacent lines of text blocks do not belong to the same text information.

It is proposed in the embodiments of this specification that the above-mentioned specific examples may be used independently, or may be used in combination for determination, which is not limited herein. In addition to the above two examples, other preset feature conditions and corresponding block features may also be analyzed or identified based on the training samples, which are not specifically limited here.

The judgment result of performing step 205 may include two types, that is, the preset characteristic condition is met or the preset characteristic condition is not met.

Step 207: Perform an operation on two adjacent lines of text blocks according to the judgment result, and the operation may include one of merging and non-merging.

Wherein, the non-merging may be arranged according to the arrangement manner corresponding to the two lines of text blocks in the original object to be recognized, or the two lines of text blocks may be isolated according to a preset strategy.

If the judgment result is merging, two adjacent lines of text blocks can be merged into a paragraph or into a file. The combined text form can be arbitrarily set without limitation.

In the embodiment of this specification, after the text blocks are combined, subsequent text processing, such as text review and quality inspection, may be performed, which is not specifically limited herein.

The embodiment of this specification provides an automatic identification scheme for whether texts are merged. This scheme can automatically identify whether two adjacent lines of text blocks belong to the same text block by using the block features of the text blocks and the preset feature conditions established in the training samples. text information, and determine whether to merge the two adjacent lines of text blocks according to the judgment result. The solutions described in the embodiments of this specification can improve text recognition efficiency.

As shown in Figure 3, "×" represents text, and Figure 3 shows two lines of text blocks adjacent to each other, and the line height far exceeds the line spacing. In this case, it can be determined that the two lines of text blocks belong to the same text information, and can merge together.

As shown in FIG. 4 , the two lines of text blocks adjacent to the top and bottom are relatively close in line height and line spacing. In this case, it can be determined that the two lines of text blocks belong to the same text information and can be merged together.

As shown in FIG. 5 , in two adjacent lines of text blocks, if the line height of the upper line of text blocks is greater than that of the lower line of text blocks, it is determined that the two lines of text blocks do not belong to the same text information and need not be merged.

FIG. 6 is a flowchart of an application example of a text recognition method proposed in an embodiment of the present specification.

Step 602: Detect the service information published by the user.

The execution body of the method can be a business information platform, which can receive various business information uploaded or released by the user, that is, detect the business information released by the user. The types of business information may be videos, texts, pictures, texts such as posters, pictorials, etc., which are not specifically limited here.

In the embodiment of this specification, detecting the service information released by the user may include regularly or irregularly detecting the service information released by the user, and may also be the service information obtained by retrieving and scanning from a database, which is not specifically limited here.

Step 604: Extract the object to be identified from the service information.

The service information recorded in the embodiments of this specification may include objects to be identified, such as images. In this case, the business information can be text containing pictures, or it can be a video, and the image is an image frame extracted from the video.

Step 606 can refer to the content of step 201 above, step 608 can refer to the content of step 203 above, step 610 can refer to the content of step 205 above, and step 612 can refer to the content of step 207 above, which is not specifically limited here.

FIG. 7 is a flowchart of a text recognition method proposed in an embodiment of the present specification. This method describes a method of forming a merged model that is applied in practice.

Step 702: Perform text recognition on the training sample, and obtain text blocks based on the recognized lines of text.

For the text recognition scheme of the training samples, reference may be made to step 201 above, which will not be repeated here.

Among them, the training samples may include one or both of white samples and black samples. The so-called two adjacent lines of text in the white sample belong to the same text information and can be merged together, while the two adjacent lines of text in the black sample belong to the same text information. Blocks cannot be merged. The sample size of the white sample and the black sample can be selected according to actual needs and is not used as a specific limitation.

Step 704: Extract block features of the text block.

For details, reference may be made to step 203 above, which will not be repeated here.

Step 706: Use the block feature of the text block to train the merging model to determine the preset feature conditions in the merging model, so that when identifying the block features of two adjacent lines of text blocks in the object to be recognized, judge the two adjacent lines. Whether the block feature of the text block meets the preset feature condition, an operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The embodiments of this specification model the process of judging whether the block features of the text blocks in two adjacent lines meet the preset feature conditions, and train the merging model based on the principle of machine learning, so that the merging model is based on the The block feature of the text block continuously adjusts and updates the preset feature conditions until the model result of the merged model reaches the preset effect.

The merging model described in the embodiments of the present specification may select a classification model, such as a decision tree, etc., which is not specifically limited here.

In addition, before using the block feature of the text block to train the merged model, the method may further include:

Acquiring merge information of the text blocks of two adjacent lines in the training sample, where the merge information represents whether the two adjacent text blocks are merged;

The block features are marked according to the merge information.

The merge information records whether the two adjacent lines of text blocks can be merged, and based on this, tags are added to the block features to form a mark. This embodies a supervised learning method and improves the training efficiency of the merged model.

In another embodiment, an unsupervised learning method or a semi-supervised learning method may be used to train the merged model without pre-marking.

In other embodiments of the present specification, a manual identification scheme may also be used to determine the preset characteristic conditions.

FIG. 8 is a flowchart of a text recognition method proposed in an embodiment of the present specification.

For step 801, reference may be made to the content of step 201 above, and for step 803, reference may be made to the content of step 203 above, which is not specifically limited herein.

Step 805: Use the merge model to process the block features of the two adjacent lines of text blocks to obtain a judgment result of whether the adjacent two lines of text blocks meet the preset feature conditions. The merge model is to use the two adjacent lines identified from the training samples. The block features of the text blocks are trained to determine the preset feature conditions.

The block features of two adjacent lines of text blocks are processed by using the merging model, including:

Taking the block features of two adjacent lines of text blocks as input, the merging model is used to judge whether the block features of the two adjacent lines of text blocks meet the preset feature conditions, and then the judgment result is obtained.

For step 807, reference may be made to the above step 207, which is not specifically limited here.

FIG. 9 is a flowchart of an application example of a text recognition method proposed in an embodiment of the present specification.

Step 902: Acquire training pictures.

Step 904: perform OCR identification on the training picture;

Step 906: combine each line of text identified to obtain a text block, and obtain the coordinates of each text block;

Step 908: Calculate the line height of the text block according to the coordinates of the text block;

Step 910: Calculate the line spacing between the two adjacent lines of text blocks according to the coordinates of the two adjacent lines of text blocks;

Step 912: Mark whether two adjacent lines of text blocks are merged;

Step 914: Obtain the block feature and the marked merge information composed of the line height of the text block and the line spacing between two adjacent lines of text blocks;

Step 916: Train a decision tree model by using the acquired block features and the marked merge information to adjust the preset feature conditions to obtain a merge model.

FIG. 10 is a flowchart of a text recognition method proposed in an embodiment of the present specification.

Step 1001: Obtain a predicted picture.

Step 1003: perform OCR identification on the predicted picture;

Step 1005: combine each line of text identified to obtain a text block, and obtain the coordinates of each text block;

Step 1007: Calculate the line height of the text block according to the coordinates of the text block;

Step 1009: Calculate the line spacing between the two lines of text blocks according to the coordinates of the two adjacent lines of text blocks;

Step 1011: Mark whether two adjacent lines of text blocks are merged;

Step 1013: Obtain the block feature formed by the line height of the text block and the line spacing between two adjacent lines of text blocks;

Step 1015: Input the acquired block feature into the merging model obtained by training, and determine whether the block feature reaches the preset feature condition by the merging model;

Step 1017 : whether the merged model outputs the merged result.

FIG. 11 is a schematic structural diagram of a text recognition apparatus according to an embodiment of the present specification.

The device may include:

The text recognition module 1101 is to perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

Extraction module 1102, extracting the block feature of the text block;

The judgment module 1103 judges whether the block features of the text blocks in two adjacent lines meet a preset feature condition, and the preset feature condition is established by using the training sample when the two adjacent lines of text blocks belong to the same text information, The feature conditions satisfied by the block features of the two adjacent lines of text blocks;

The execution module 1104 performs an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

The text recognition device described in the embodiments of this specification provides an automatic recognition scheme for whether texts are merged or not, and can automatically recognize two adjacent lines of text blocks by using the block features of the text blocks and the preset feature conditions established in the training samples. Whether it belongs to the same text information, and whether to merge the two adjacent lines of text blocks is determined according to the judgment result. The solutions described in the embodiments of this specification can improve text recognition efficiency.

Based on the same inventive concept, the embodiments of this specification also provide an electronic device, including:

processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

It is judged whether the block features of the text blocks in two adjacent lines meet a preset feature condition, and the preset feature condition is that when two adjacent lines of text blocks belong to the same text information established by using training samples, the adjacent text blocks belong to the same text information. The feature conditions satisfied by the block feature of the two-line text block;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

Based on the same inventive concept, the embodiments of this specification also provide a computer-readable storage medium, including a computer program used in combination with an electronic device, and the computer program can be executed by a processor to complete the following steps:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

It is judged whether the block features of the text blocks in two adjacent lines meet a preset feature condition, and the preset feature condition is that when two adjacent lines of text blocks belong to the same text information established by using training samples, the adjacent text blocks belong to the same text information. The feature conditions satisfied by the block feature of the two-line text block;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

FIG. 12 is a schematic structural diagram of a text recognition apparatus according to an embodiment of the present specification.

The device may include:

The text recognition module 1201 performs text recognition on the training samples, and obtains text blocks based on the recognized lines of text;

Extraction module 1202, extracts the block feature of the text block;

The training module 1203 uses the block features of the text blocks to train the merging model to determine the preset feature conditions in the merging model, so that when identifying the block features of two adjacent lines of text blocks in the object to be recognized, determine the Whether the block feature of the text blocks in two adjacent lines meets the preset feature condition, an operation is performed on the text blocks in the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

Based on the same inventive concept, the embodiments of this specification also provide an electronic device, including:

processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

Perform text recognition on the training samples, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

The merging model is trained by using the block features of the text blocks to determine the preset feature conditions in the merging model, so that when the block features of two adjacent lines of text blocks in the object to be recognized are identified, the whether the block feature of the text block meets the preset feature condition, and perform an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and not merging.

Based on the same inventive concept, the embodiments of this specification also provide a computer-readable storage medium, including a computer program used in combination with an electronic device, and the computer program can be executed by a processor to complete the following steps:

Perform text recognition on the training samples, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

The merging model is trained by using the block features of the text blocks to determine the preset feature conditions in the merging model, so that when the block features of two adjacent lines of text blocks in the object to be recognized are identified, the whether the block feature of the text block meets the preset feature condition, and perform an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and not merging.

FIG. 13 is a schematic structural diagram of a text recognition device according to an embodiment of the present specification

The device may include:

The text recognition module 1301 is to perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

Extraction module 1302, extracts the block feature of the text block;

The model processing module 1303 uses the merging model to process the block features of the text blocks in the two adjacent lines, and obtains the judgment result of whether the text blocks in the two adjacent lines meet the preset feature conditions. The merging model uses Obtained by training the block features of two adjacent lines of text blocks identified in the training samples to determine the preset feature conditions;

The execution module 1304 performs an operation on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

Based on the same inventive concept, the embodiments of this specification also provide an electronic device, including:

processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

The block features of the text blocks in the two adjacent lines are processed by the merging model, and the judgment result of whether the text blocks in the adjacent two lines meet the preset feature conditions is obtained. The block features of the two-line text blocks are trained to determine the preset feature conditions;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

Based on the same inventive concept, the embodiments of this specification also provide a computer-readable storage medium, including a computer program used in combination with an electronic device, and the computer program can be executed by a processor to complete the following steps:

Perform text recognition on the object to be recognized, and obtain text blocks based on the recognized lines of text;

extracting block features of the text block;

A merging model is used to process the block features of the text blocks in two adjacent lines to obtain a judgment result of whether the text blocks in the two adjacent lines meet the preset feature conditions. The block features of the two adjacent lines of text blocks obtained are obtained by training to determine the preset feature conditions;

An operation is performed on the text blocks of the two adjacent lines according to the judgment result, and the operation includes one of merging and non-merging.

In the 1990s, improvements in a technology could be clearly differentiated between improvements in hardware (eg, improvements to circuit structures such as diodes, transistors, switches, etc.) or improvements in software (improvements in method flow). However, with the development of technology, the improvement of many methods and processes today can be regarded as a direct improvement of the hardware circuit structure. Designers almost get the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that the improvement of a method flow cannot be realized by hardware entity modules. For example, a Programmable Logic Device (PLD) (eg, Field Programmable Gate Array (FPGA)) is an integrated circuit whose logic function is determined by user programming of the device. It is programmed by the designer to "integrate" a digital system on a PLD without having to ask the chip manufacturer to design and manufacture a dedicated integrated circuit chip. And, instead of making integrated circuit chips by hand, these days, much of this programming is done using software called a "logic compiler", which is similar to the software compiler used in program development and writing, but before compiling The original code also has to be written in a specific programming language, which is called Hardware Description Language (HDL), and there is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language) , AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (RubyHardware Description Language), etc. The most commonly used are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. It should also be clear to those skilled in the art that a hardware circuit for implementing the logic method process can be easily obtained by simply programming the method process in the above-mentioned several hardware description languages and programming it into the integrated circuit.

The controller may be implemented in any suitable manner, for example, the controller may take the form of eg a microprocessor or processor and a computer readable medium storing computer readable program code (eg software or firmware) executable by the (micro)processor , logic gates, switches, application specific integrated circuits (ASICs), programmable logic controllers and embedded microcontrollers, examples of controllers include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicon Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that, in addition to implementing the controller in the form of pure computer-readable program code, the controller can be implemented as logic gates, switches, application-specific integrated circuits, programmable logic controllers and embedded devices by logically programming the method steps. The same function can be realized in the form of a microcontroller, etc. Therefore, such a controller can be regarded as a hardware component, and the devices included therein for realizing various functions can also be regarded as a structure within the hardware component. Or even, the means for implementing various functions can be regarded as both a software module implementing a method and a structure within a hardware component.

The systems, devices, modules or units described in the above embodiments may be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or A combination of any of these devices.

For the convenience of description, when describing the above device, the functions are divided into various units and described respectively. Of course, when implementing the present application, the functions of each unit may be implemented in one or more software and/or hardware.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable data processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

Memory may include non-persistent memory in computer readable media, random access memory (RAM) and/or non-volatile memory in the form of, for example, read only memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media includes both persistent and non-permanent, removable and non-removable media, and storage of information may be implemented by any method or technology. Information may be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Flash Memory or other memory technology, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

It should also be noted that the terms "comprising", "comprising" or any other variation thereof are intended to encompass a 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 expressly listed, or which are inherent to such a process, method, article of manufacture, or apparatus are also included. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article of manufacture, or device that includes the element.

The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including storage devices.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, as for the system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for related parts, please refer to the partial descriptions of the method embodiments.

The above descriptions are merely examples of the present application, and are not intended to limit the present application. Various modifications and variations of this application are possible for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the scope of the claims of this application.

Claims (17)

1. A text recognition method, comprising:

carrying out character recognition on an object to be recognized, and respectively obtaining text blocks based on recognized characters of each line;

extracting block features of the text block;

judging whether the block features of two adjacent lines of text blocks reach a preset feature condition, wherein the preset feature condition is a feature condition which is established by using a training sample and is met by the block features of the two adjacent lines of text blocks when the two adjacent lines of text blocks belong to the same text information;

and executing operation on the two adjacent lines of text blocks according to the judgment result, wherein the operation comprises one of merging and non-merging.

2. The method of claim 1, the object to be identified being an image.

3. The method of claim 1, prior to performing text recognition on the object to be recognized, further comprising:

detecting service information issued by a user;

and extracting the object to be identified from the service information.

4. The method of claim 1, wherein the block characteristics include one or both of a line height of each of the text blocks and a line spacing between two adjacent lines of the text blocks.

5. The method of claim 4, wherein determining whether the block features of two adjacent lines of the text block meet a preset feature condition comprises:

and judging that the line height of the text blocks in the two adjacent lines is not less than the line spacing.

6. The method of claim 5, wherein determining that the line height of the two adjacent lines of the text block is not less than the line spacing comprises:

and judging whether the line height of the two adjacent lines of text blocks exceeds the line spacing to reach a preset difference value.

7. The method of claim 4, wherein determining whether the block features of two adjacent lines of the text block meet a preset feature condition comprises:

and judging whether the difference between the line heights of the two adjacent lines of text blocks is not greater than a preset line height difference.

8. The method of claim 4, wherein performing an operation on the two adjacent lines of text blocks according to the determination result comprises:

and if the judgment result is that the preset characteristic condition is met, combining the two adjacent lines of text blocks into a paragraph.

9. A text recognition method, comprising:

carrying out character recognition on the training sample, and respectively obtaining text blocks based on recognized characters of each line;

extracting block features of the text block;

and training a merging model by using the block features of the text blocks to determine a preset feature condition in the merging model, so that when the block features of two adjacent lines of text blocks in the object to be recognized are recognized, whether the block features of the two adjacent lines of text blocks reach the preset feature condition is judged, and operation is performed on the two adjacent lines of text blocks according to a judgment result, wherein the operation comprises one of merging and non-merging.

10. The method of claim 9, prior to training a merging model using block features of the text block, further comprising:

acquiring merging information of two adjacent lines of the text blocks in the training sample, wherein the merging information represents whether the two adjacent text blocks are merged or not;

and marking the block features according to the merging information.

11. A text recognition method, comprising:

carrying out character recognition on an object to be recognized, and respectively obtaining text blocks based on recognized characters of each line;

extracting block features of the text block;

processing the block features of two adjacent lines of text blocks by using a merging model to obtain a judgment result of whether the two adjacent lines of text blocks reach a preset feature condition, wherein the merging model is obtained by training by using the block features of two adjacent lines of text blocks identified from a training sample to determine the preset feature condition;

and executing operation on the two adjacent lines of text blocks according to the judgment result, wherein the operation comprises one of merging and non-merging.

12. A text recognition apparatus comprising:

the character recognition module is used for carrying out character recognition on the object to be recognized and respectively obtaining text blocks based on recognized characters of each row;

the extraction module is used for extracting block features of the text block;

the judging module is used for judging whether the block features of the two adjacent lines of text blocks reach a preset feature condition, wherein the preset feature condition is a feature condition which is established by using a training sample and is met by the block features of the two adjacent lines of text blocks when the two adjacent lines of text blocks belong to the same text information;

and the execution module executes operation on the two adjacent lines of text blocks according to the judgment result, wherein the operation comprises one of combination and non-combination.

13. A text recognition apparatus comprising:

the character recognition module is used for carrying out character recognition on the training sample and respectively obtaining text blocks based on recognized characters of each line;

the extraction module is used for extracting block features of the text block;

and the training module is used for training a merging model by using the block features of the text blocks so as to determine a preset feature condition in the merging model, so that when the block features of two adjacent lines of text blocks in the object to be recognized are recognized, whether the block features of the two adjacent lines of text blocks reach the preset feature condition is judged, and operation is performed on the two adjacent lines of text blocks according to a judgment result, wherein the operation comprises one of merging and non-merging.

14. A text recognition apparatus comprising:

the character recognition module is used for carrying out character recognition on the object to be recognized and respectively obtaining text blocks based on recognized characters of each row;

the extraction module is used for extracting block features of the text block;

the model processing module is used for processing the block characteristics of two adjacent lines of text blocks by utilizing a merging model to obtain a judgment result of whether the two adjacent lines of text blocks reach a preset characteristic condition, and the merging model is obtained by utilizing the block characteristics of two adjacent lines of text blocks identified from a training sample to train so as to determine the preset characteristic condition;

and the execution module executes operation on the two adjacent lines of text blocks according to the judgment result, wherein the operation comprises one of combination and non-combination.

15. An electronic device, comprising:

a processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

carrying out character recognition on an object to be recognized, and respectively obtaining text blocks based on recognized characters of each line;

extracting block features of the text block;

judging whether the block features of two adjacent lines of text blocks reach a preset feature condition, wherein the preset feature condition is a feature condition which is established by using a training sample and is met by the block features of the two adjacent lines of text blocks when the two adjacent lines of text blocks belong to the same text information;

and executing operation on the two adjacent lines of text blocks according to the judgment result, wherein the operation comprises one of merging and non-merging.

16. An electronic device, comprising:

a processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

carrying out character recognition on the training sample, and respectively obtaining text blocks based on recognized characters of each line;

extracting block features of the text block;

and training a merging model by using the block features of the text blocks to determine a preset feature condition in the merging model, so that when the block features of two adjacent lines of text blocks in the object to be recognized are recognized, whether the block features of the two adjacent lines of text blocks reach the preset feature condition is judged, and operation is performed on the two adjacent lines of text blocks according to a judgment result, wherein the operation comprises one of merging and non-merging.

17. An electronic device, comprising:

a processor; and

a memory configured to store a computer program that, when executed, causes the processor to:

carrying out character recognition on an object to be recognized, and respectively obtaining text blocks based on recognized characters of each line;

extracting block features of the text block;

processing the block features of two adjacent lines of text blocks by using a merging model to obtain a judgment result of whether the two adjacent lines of text blocks reach a preset feature condition, wherein the merging model is obtained by training by using the block features of two adjacent lines of text blocks identified from a training sample to determine the preset feature condition;

and executing operation on the two adjacent lines of text blocks according to the judgment result, wherein the operation comprises one of merging and non-merging.

Images