A method for data recognition and an intelligent interactive display device

Abstract

The application discloses a data recognition method and an intelligent interactive display device, which are used for distinguishing the content written by a user, avoiding recognizing the whole content, supporting the user in modifying the scene locally, and improving the accuracy of recognition. The method comprises the following steps: receiving first handwriting information written by a user; determining second handwriting information bound with at least one cell from the first handwriting information according to the position relationship between each cell in the generated table and the first handwriting information; and recognizing the second handwriting information bound with the at least one cell to obtain recognition data corresponding to the at least one cell.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a data recognition method and an intelligent interactive display device. BACKGROUND

[0002] An intelligent conference is an office product integrating electronic whiteboards, computers, remote conference terminals and other devices, and has multiple functions such as interactive intelligent whiteboards. With the progress of high-precision positioning technologies such as infrared, electromagnetic and capacitive touch, electronic whiteboards can now realize touch writing. During the writing process on the electronic whiteboard, it is often necessary to present the array on the electronic whiteboard in the form of a chart, and generate a chart corresponding to the data drawn by the user on the electronic whiteboard.

[0003] However, the current recognition of user-drawn data has many defects. Usually, the entire content written by the user on the electronic whiteboard is recognized, which has low recognition accuracy and poor user experience. SUMMARY

[0004] The present application provides a data recognition method and an intelligent interactive display device, which are used to distinguish the content written by the user, avoid recognizing the entire content, support the user's local modification scenario, and improve the recognition accuracy.

[0005] In a first aspect, the present application provides a data recognition method, which includes:

[0006] receiving first handwriting information written by a user;

[0007] determining, from the first handwriting information, second handwriting information bound to at least one cell according to the positional relationship between the cells in the generated table and the first handwriting information;

[0008] recognizing the second handwriting information bound to the at least one cell to obtain recognition data corresponding to the at least one cell.

[0009] The data recognition method provided in this embodiment uses the binding relationship between the cells in the generated table and the first handwriting information written by the user to separately recognize the second handwriting information bound to the cells, thereby avoiding recognizing the entire content written by the user at once. Based on the binding relationship with each cell, the array written by the user is determined, and the row-column relationship is determined according to the binding relationship, so that an accurate table is drawn. Since this embodiment supports local recognition, it can also be applied to the user's local modification scenario, separately recognizes the local modification, improves the recognition efficiency and accuracy, and improves the user's use experience.

[0010] As an optional implementation,

[0011] The table is generated according to the first handwriting information written by the user; or

[0012] The table is generated according to preset row and column parameters.

[0013] As an optional implementation, the receiving of the first handwriting information written by the user comprises:

[0014] Displaying the generated table in a display area;

[0015] Receiving first handwriting information written by the user in a cell of the displayed table.

[0016] As an optional implementation, the determining of second handwriting information bound to at least one cell from the first handwriting information according to the positional relationship between each cell in the generated table and the first handwriting information comprises:

[0017] When the first handwriting information is located within the cell, the first handwriting information is determined as the second handwriting information bound to the cell; or

[0018] When the landing point of the first handwriting information is located within the cell, the first handwriting information is determined as the second handwriting information bound to the cell; or

[0019] When the center of gravity of the first handwriting information is located within the cell, the first handwriting information is determined as the second handwriting information bound to the cell.

[0020] As an optional implementation, the size of the table is adaptively adjusted according to the size of the second handwriting information bound to the cell.

[0021] As an optional implementation, the method further comprises:

[0022] When the second handwriting information bound to the cell exceeds the area covered by the cell, the row and column where the cell is located are expanded so that the second handwriting information is located within the area covered by the cell.

[0023] As an optional implementation, the method further comprises:

[0024] Determining the offset of at least one cell in the table except the expanded cell, wherein the offset of the cell is based on the expansion of the cells in the table;

[0025] According to the offset amount of the at least one offset cell, the second handwriting information bound to the at least one offset cell is correspondingly offset.

[0026] As an optional implementation, after the identification data corresponding to the cell is obtained, the method further comprises:

[0027] Converting the identification data into standard data conforming to a preset specification;

[0028] Scaling the standard data according to the size of the area covered by the cell, so that the standard data is located in the area covered by the cell.

[0029] As an optional implementation, the method further comprises:

[0030] Identifying the second handwriting information bound to the cell in the table at least twice.

[0031] As an optional implementation, the identifying the second handwriting information bound to the cell in the table at least twice comprises:

[0032] When it is determined that the second handwriting information bound to the cell has changed after the last identification of the second handwriting information bound to the cell, identifying the second handwriting information bound to the cell this time.

[0033] As an optional implementation, the determining that the second handwriting information bound to the cell has changed comprises:

[0034] Comparing the second handwriting information bound to the cell before this identification and the second handwriting information bound to the cell at the last identification;

[0035] When the comparison result is inconsistent, it is determined that the second handwriting information bound to the cell has changed.

[0036] As an optional implementation, the identifying the second handwriting information bound to the cell this time when the comparison result is inconsistent comprises:

[0037] When the comparison result is inconsistent, the second handwriting information bound to the cell in the table is identified this time by using a second identification algorithm;

[0038] Wherein, the first identification algorithm is used at the last identification.

[0039] As an optional implementation,

[0040] The first identification algorithm comprises a recurrent neural network algorithm; and the second identification algorithm comprises a convolutional neural network algorithm.

[0041] As an optional implementation,

[0042] The changed second handwriting information is obtained by at least one of modification, deletion and addition of the second handwriting information of the cell binding after the user recognizes the second handwriting information of the cell binding last time.

[0043] As an optional implementation, the recognizing the second handwriting information of the at least one cell binding to obtain the corresponding recognition data of the at least one cell includes:

[0044] When the plurality of cells in the table are all bound with the second handwriting information, the second handwriting information of the plurality of cells is recognized in parallel;

[0045] According to the row and column information of the cell bound with the second handwriting information in the table, the corresponding relationship between the recognition data obtained by recognition and the plurality of cells is determined to obtain the corresponding recognition data of the plurality of cells.

[0046] As an optional implementation, the method further includes:

[0047] After the corresponding recognition data of the plurality of cells is obtained, a chart is generated according to the corresponding recognition data of the plurality of cells and the row and column information of each cell in the table.

[0048] As an optional implementation, the transparency of the table display is adjustable.

[0049] As an optional implementation, the table is not displayed in the display area.

[0050] In a second aspect, an intelligent interactive display device is provided, which includes a display screen and a control circuit, and wherein:

[0051] The display screen is configured to display content.

[0052] The control circuit includes a processor and a memory, the memory is configured to store a program executable by the processor, and the processor is configured to read the program in the memory and perform the following steps:

[0053] Receiving first handwriting information written by a user;

[0054] According to the position relationship between each cell in the generated table and the first handwriting information, determining second handwriting information bound with at least one cell from the first handwriting information;

[0055] Recognizing the second handwriting information bound with at least one cell to obtain corresponding recognition data of the at least one cell.

[0056] As an optional implementation,

[0057] The table is generated according to the first handwriting information written by the user; or

[0058] The table is generated according to preset row and column parameters.

[0059] As an optional implementation, the processor is specifically configured to perform:

[0060] displaying the generated table in a display area;

[0061] receiving first handwriting information written by the user in a cell of the displayed table.

[0062] As an optional implementation, the processor is specifically configured to perform:

[0063] determining the first handwriting information as second handwriting information bound to the cell when the first handwriting information is located within the cell; or

[0064] determining the first handwriting information as second handwriting information bound to the cell when a landing point of the first handwriting information is located within the cell; or

[0065] determining the first handwriting information as second handwriting information bound to the cell when a center of gravity of the first handwriting information is located within the cell.

[0066] As an optional implementation, a size of the table is adaptively adjusted according to a size of the second handwriting information bound to the cell.

[0067] As an optional implementation, the processor is specifically further configured to perform:

[0068] when the second handwriting information bound to the cell exceeds an area covered by the cell, expanding a row and a column where the cell is located so that the second handwriting information is located within the area covered by the cell.

[0069] As an optional implementation, the processor is specifically further configured to perform:

[0070] determining at least one offset cell in the table except the expanded cell, wherein the offset of the cell is based on the expansion of the cell in the table;

[0071] offsetting second handwriting information bound to the at least one offset cell according to an offset amount of the at least one offset cell.

[0072] As an optional implementation, after obtaining the identification data corresponding to the cell, the processor is specifically further configured to perform:

[0073] convert the identification data into standard data conforming to a preset specification;

[0074] scale the standard data according to the size of the area covered by the cell, so that the standard data is located in the area covered by the cell.

[0075] As an optional implementation, the processor is specifically further configured to perform:

[0076] identify the second handwriting information bound to the cell in the table at least twice.

[0077] As an optional implementation, the processor is specifically configured to perform:

[0078] when it is determined that the second handwriting information bound to the cell has changed after the last identification of the second handwriting information bound to the cell, identify the second handwriting information bound to the cell this time.

[0079] As an optional implementation, the processor is specifically configured to perform:

[0080] compare the second handwriting information bound to the cell before this identification and the second handwriting information bound to the cell at the last identification;

[0081] when the comparison result is inconsistent, determine that the second handwriting information bound to the cell has changed.

[0082] As an optional implementation, the processor is specifically configured to perform:

[0083] when the comparison result is inconsistent, identify the second handwriting information bound to the cell in the table this time by using a second identification algorithm;

[0084] wherein the first identification algorithm is used at the last identification.

[0085] As an optional implementation,

[0086] the first identification algorithm includes a recurrent neural network algorithm; and the second identification algorithm includes a convolutional neural network algorithm.

[0087] As an optional implementation,

[0088] The changed second handwriting information is obtained by at least one of modification, deletion and addition of the second handwriting information of the cell binding by the user after the last recognition of the second handwriting information of the cell binding.

[0089] As an optional implementation, the processor is specifically configured to perform:

[0090] When the plurality of cells in the table are all bound with the second handwriting information, the second handwriting information of the plurality of cells is recognized in parallel;

[0091] According to the row and column information of the cell bound with the second handwriting information in the table, a corresponding relationship between the recognition data obtained by recognition and the plurality of cells is determined, and the recognition data corresponding to the plurality of cells is obtained.

[0092] As an optional implementation, the processor is specifically configured to perform:

[0093] When the recognition data corresponding to the plurality of cells is obtained, a graph is generated according to the recognition data corresponding to the plurality of cells and the row and column information of each cell in the table.

[0094] As an optional implementation, the transparency of the table display is adjustable.

[0095] As an optional implementation, the table is not displayed in the display area.

[0096] In a third aspect, an embodiment of the present application further provides a data recognition device, comprising:

[0097] A receiving unit is configured to receive first handwriting information written by a user;

[0098] A binding unit is configured to determine second handwriting information bound with at least one cell from the first handwriting information according to a position relationship between each cell in a generated table and the first handwriting information;

[0099] An identification unit is configured to identify the second handwriting information bound with the at least one cell, and obtain recognition data corresponding to the at least one cell.

[0100] As an optional implementation,

[0101] The table is generated according to the first handwriting information written by the user; or,

[0102] The table is generated according to preset row and column parameters.

[0103] As an optional implementation, the receiving unit is specifically configured to:

[0104] displaying the generated table in a display area;

[0105] receiving first handwriting information written by a user in a cell of the displayed table.

[0106] As an optional implementation, the binding unit is specifically configured to:

[0107] determine the first handwriting information as second handwriting information bound to the cell when the first handwriting information is located within the cell; or,

[0108] determine the first handwriting information as second handwriting information bound to the cell when a landing point of the first handwriting information is located within the cell; or,

[0109] determine the first handwriting information as second handwriting information bound to the cell when a center of gravity of the first handwriting information is located within the cell.

[0110] As an optional implementation, a size of the table is adaptively adjusted according to a size of the second handwriting information bound to the cell.

[0111] As an optional implementation, the method further includes an expanding unit specifically configured to:

[0112] expand a row and a column in which the cell is located when the second handwriting information bound to the cell exceeds an area covered by the cell, so that the second handwriting information is located within the area covered by the cell.

[0113] As an optional implementation, the method further includes an offsetting unit specifically configured to:

[0114] determine offsets of at least one cell other than the expanded cell in the table, wherein the offsets of the cell are based on the cell after expansion in the table;

[0115] offset the second handwriting information bound to the at least one offset cell according to the offsets of the at least one offset cell.

[0116] As an optional implementation, after the identification data corresponding to the cell is obtained, the method further includes a scaling unit specifically configured to:

[0117] convert the identification data into standard data conforming to a preset specification;

[0118] scale the standard data according to a size of an area covered by the cell, so that the standard data is located within the area covered by the cell.

[0119] As an optional implementation, the re-identification unit is specifically configured to:

[0120] The second handwriting information of the cell binding in the table is identified at least twice.

[0121] As an optional implementation, the re-identification unit is specifically configured to:

[0122] When it is determined that the second handwriting information of the cell binding has changed after the last identification of the second handwriting information of the cell binding, the second handwriting information of the cell binding is identified this time.

[0123] As an optional implementation, the re-identification unit is specifically configured to:

[0124] The second handwriting information of the cell binding before this time of identification and the second handwriting information of the cell binding at the last time of identification are compared;

[0125] When the comparison result is inconsistent, it is determined that the second handwriting information of the cell binding has changed.

[0126] As an optional implementation, the re-identification unit is specifically configured to:

[0127] When the comparison result is inconsistent, the second handwriting information of the cell binding in the table is identified this time by using a second identification algorithm;

[0128] Wherein, the first identification algorithm is used at the last time of identification.

[0129] As an optional implementation,

[0130] The first identification algorithm includes a recurrent neural network algorithm; and the second identification algorithm includes a convolutional neural network algorithm.

[0131] As an optional implementation,

[0132] The changed second handwriting information is obtained by at least one of modification, deletion, and addition of the second handwriting information of the cell binding by a user after the last identification of the second handwriting information of the cell binding.

[0133] As an optional implementation, the identification unit is specifically configured to:

[0134] When multiple cells in the table are all bound with the second handwriting information, the second handwriting information of the multiple cells is identified in parallel;

[0135] According to the row and column information of the cell bound with the second handwriting information in the table, the correspondence between the recognized data and the plurality of cells is determined, and the recognized data corresponding to the plurality of cells is obtained.

[0136] As an optional implementation, the chart generating unit is specifically configured to:

[0137] When the recognized data corresponding to the plurality of cells is obtained, a chart is generated according to the recognized data corresponding to the plurality of cells and the row and column information of each cell in the table.

[0138] As an optional implementation, the transparency of the table display is adjustable.

[0139] As an optional implementation, the table is not displayed in the display area.

[0140] In a fourth aspect, the embodiments of the present application further provide a computer storage medium, which stores a computer program, and the program is executed by a processor to implement the steps of the method in the first aspect.

[0141] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0142] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0143] Figure 1 A data recognition method implementation flowchart is provided for the embodiments of the present application;

[0144] Figure 2 A table self-adaptive expansion schematic diagram is provided for the embodiments of the present application;

[0145] Figure 3 A handwriting offset schematic diagram after table expansion is provided for the embodiments of the present application;

[0146] Figure 4 A table size fixed adaptive adjustment content schematic diagram is provided for the embodiments of the present application;

[0147] Figure 5 A chart drawing data based recognition method implementation flowchart is provided for the embodiments of the present application;

[0148] Figure 6A recognition method flow chart based on chart drawing data provided by the embodiment of the present application is provided;

[0149] Figure 7 A schematic diagram of an intelligent interactive display device provided by the embodiment of the present application is provided;

[0150] Figure 8 A schematic diagram of a data recognition device provided by the embodiment of the present application is provided. DETAILED DESCRIPTION

[0151] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0152] In the embodiments of the present application, the term "and / or" describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can mean that there are three cases of A alone, A and B together, and B alone. The character " / " generally represents that the associated objects before and after it are in an "or" relationship.

[0153] The application scenarios described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems. In the description of the present application, unless otherwise specified, the meaning of "multiple" is two or more.

[0154] Embodiment 1, the intelligent conference is an office product integrating electronic whiteboard, computer and remote conference terminal and various devices, and has various functions such as interactive intelligent whiteboard. With the progress of high-precision positioning technologies such as infrared, electromagnetic and capacitive touch, the electronic whiteboard can realize touch writing. In the process of writing on the electronic whiteboard, it is often necessary to present the array on the electronic whiteboard in the form of a chart, and generate a chart corresponding to the data by the data drawn by the user on the electronic whiteboard. However, the current recognition of the data drawn by the user has many defects. Usually, the entire content written by the user on the electronic whiteboard is recognized. Due to the user's handwriting error, it is difficult to ensure that the arrays are aligned. When the arrays in the same row are not in the same row, or the arrays in the same column are not in the same column, it is easy to cause the recognized arrays to be misaligned, and the finally drawn chart is also inaccurate, resulting in low recognition accuracy and poor user experience. In addition, the current recognition scheme does not support local modification, nor does it support adding content before the original content. If the user modifies an array (for example, changes 50 to 15, or changes 50 to 150) after writing all the arrays, the current scheme cannot recognize the content of the modified array, resulting in poor use effect and insufficient user experience.

[0155] In order to solve the current recognition defects of the data drawn by the user, the embodiment provides a data recognition method. The binding relationship between the generated cells in the table and the first handwriting information written by the user is used to separately recognize the second handwriting information bound to the cells, so as to avoid recognizing the entire content written by the user at one time. Based on the binding relationship with each cell, the array written by the user is determined, the row and column relationship is determined according to the binding relationship, so as to accurately draw the table. Since the embodiment supports local recognition, it can also be applied to the scene of local modification by the user, separately recognizes the local modification, improves the recognition efficiency and accuracy, and improves the user's use experience.

[0156] It should be noted that the data recognition method in the embodiment is applied to an intelligent interactive display device. The intelligent interactive display device in the embodiment refers to a large-size (generally more than 50 inches) display device with touch handwriting function. On the intelligent interactive display device, a writing program such as an electronic whiteboard is installed. The electronic whiteboard is not a traditional electronic whiteboard written with a marker or a chalk. The user can write on the intelligent display device by using a finger, an active pen or a passive pen. The intelligent interactive display device can include a liquid crystal display (LCD), an organic electroluminescence display (OLED), an electronic ink display module, and can display the handwriting in real time.

[0157] As Figure 1 shown, the embodiment provides a data recognition method, and the specific implementation process of the method is as follows:

[0158] Step 100, receiving first handwriting information written by a user;

[0159] Step 101, determining second handwriting information bound with at least one cell from the first handwriting information according to a position relationship between each cell in a generated table and the first handwriting information;

[0160] In some embodiments, the generated table in the embodiment is generated according to the first handwriting information written by the user; or, the generated table in the embodiment is generated according to preset row and column parameters.

[0161] In implementation, the first handwriting information can be divided into rows and columns according to the position of the first handwriting information written by the user on the display screen, and then the table can be generated according to the boundary lines obtained by the row and column division; or, the table can be generated according to preset row and column parameters (for example, N rows and M columns), wherein the row and column parameters include but are not limited to row height, column width, row number, column number, and the like.

[0162] In some embodiments, each cell in the table is generated in any one or more of the following ways:

[0163] Method 1) determining a column separation line according to a valley position of the first handwriting information in a first projection histogram of a first coordinate axis, determining a row separation line of any one column according to a valley position of the first handwriting information contained in the column in a second projection histogram of a second coordinate axis, and determining each cell according to the column separation line and the row separation line.

[0164] It should be noted that the first coordinate axis and the second coordinate axis in the embodiment have a vertical relationship.

[0165] In implementation, the column separation line is determined in the following way:

[0166] The column separation line is determined according to a valley position of a valley in the first projection histogram with a width greater than a threshold value.

[0167] In implementation, the row separation line is determined in the following way:

[0168] The row separation line is determined according to a valley position of a valley in the second projection histogram with a width greater than a threshold value.

[0169] In implementation, each cell of the table is generated according to the column separation line and the row separation line, specifically including:

[0170] The first handwriting information is divided into multiple columns according to the column division line, and the first handwriting information of each column is divided into multiple rows according to the row division line. Finally, the border line of the table is determined according to the minimum value and the maximum value of the handwriting points in the first handwriting information in the first coordinate axis and the second coordinate axis. The table and the data content in the table are determined according to the column division line and the row division line, the border line, and the first handwriting information. It should be noted that in the manner of generating the cells in the embodiment, the column division is performed first, and then the row division is performed on each column. The problem that the content of each row in different columns is not aligned due to the non-standard writing of the user can be solved. Even if the content belonging to different columns in each row of the user is not in the same horizontal line, the data content in each cell can still be accurately divided by using the method in the embodiment, and the division of the rows and columns is accurately realized. This implementation manner can solve the problem that the handwritten column in the vertical table is misaligned, thereby affecting the recognition.

[0171] In mode 2), the row division line is determined according to the valley position of the second projection histogram of the first handwriting information in the second coordinate axis. The column division line of any row is determined according to the valley position of the first projection histogram of the first handwriting information contained in the row. Each cell of the table is generated according to the row division line and the column division line.

[0172] In the implementation, the column division line is determined in the following manner:

[0173] The column division line is determined according to the valley position of the valley in the first projection histogram whose width is greater than a threshold value.

[0174] The row division line is determined in the following manner:

[0175] The row division line is determined according to the valley position of the valley in the second projection histogram whose width is greater than a threshold value.

[0176] In the implementation, each cell of the table is generated according to the column division line and the row division line, and specifically includes:

[0177] The first handwriting information is divided into multiple rows according to the row division line, and the first handwriting information of each row is divided into multiple columns according to the column division line. Finally, the border line of the table is determined according to the minimum value and the maximum value of the handwriting points in the first handwriting information in the first coordinate axis and the second coordinate axis. The table and the data content in the table are determined according to the column division line and the row division line, the border line, and the first handwriting information. It should be noted that in the manner of generating the cells in the embodiment, the column division is performed first, and then the row division is performed on each column. The problem that the content of each row in different columns is not aligned due to the non-standard writing of the user can be solved. Even if the content belonging to different columns in each row of the user is not in the same horizontal line, the data content in each cell can still be accurately divided by using the method in the embodiment, and the division of the rows and columns is accurately realized. This implementation manner can solve the problem that the handwritten column in the vertical table is misaligned, thereby affecting the recognition.

[0178] In the third mode, the column separation line is determined according to the valley position in the first projection histogram of the first handwriting information on the first coordinate axis, the row separation line is determined according to the valley position in the second projection histogram of the first handwriting information on the second coordinate axis, and each cell is determined according to the row separation line and the column separation line.

[0179] In an implementation, the column separation line is determined in the following manner:

[0180] The column separation line is determined according to the valley position in the first projection histogram in which the width of the valley is greater than a threshold value.

[0181] The row separation line is determined in the following manner:

[0182] The row separation line is determined according to the valley position in the second projection histogram in which the width of the valley is greater than a threshold value.

[0183] In an implementation, each cell is determined according to the column separation line and the row separation line, and specifically includes:

[0184] The first handwriting information is divided into multiple columns according to the column separation line, and is divided into multiple rows according to the row separation line. Finally, the frame line of the table is determined according to the minimum value and the maximum value of the handwriting points in the first handwriting information on the first coordinate axis and the second coordinate axis. The table and the data content in the table are determined according to the column separation line and the row separation line, the frame line, and the first handwriting information. It should be noted that in the manner of generating the cell in this embodiment, the division of the rows and columns is simultaneously implemented, which can solve the problem that the empty cells appear in the table due to the non-standard writing of the user. By using the method in this embodiment, the empty cells, i.e., the cells without the first handwriting information, can be determined, the data content in each cell is accurately divided, and the division of the rows and columns is accurately implemented.

[0185] In some embodiments, the first projection histogram is obtained by performing a down-sampling on a projection histogram of the first handwriting information projected onto a first coordinate axis, wherein the down-sampling is used to reduce a unit interval of the first coordinate axis of the projection histogram to a first preset interval, and sum up the number of handwriting points projected in the first preset interval, and the sum is taken as the number of handwriting points corresponding to the first preset interval; for example, taking the first coordinate axis as the X axis, the X axis coordinate of the projection histogram represents the position of the handwriting point in the first handwriting information in the X direction, and the Y axis coordinate represents the number of handwriting points projected onto different X axis positions; for example, the number of handwriting points in the projection histogram in the 0-100 interval of the X axis is down-sampled to the new 0-10 interval, specifically, the sum of the number of handwriting points in the 0-10 sub-interval in the 0-100 interval is taken as the number of handwriting points corresponding to the 0-1 sub-interval in the new 0-10 interval, the sum of the number of handwriting points in the 10-20 sub-interval in the 0-100 interval is taken as the number of handwriting points corresponding to the 1-2 sub-interval in the new 0-10 interval, and so on. Further, the number of handwriting points in each sub-interval in the 0-10 interval can be normalized, for example, taking the 0-1 sub-interval in the 0-10 interval, the sum of the number of handwriting points corresponding to the 0-1 sub-interval is taken as the number of handwriting points corresponding to the 0-1 sub-interval, wherein the 0-1 sub-interval corresponds to a number of handwriting points, i.e., the 0-1 sub-interval corresponds to a value.

[0186] The second projection histogram is obtained by performing a down-sampling on a projection histogram of the first handwriting information projected onto a second coordinate axis, wherein the down-sampling is used to reduce a unit interval of the second coordinate axis of the projection histogram to a second preset interval, and sum up the number of handwriting points projected in the second preset interval, and the sum is taken as the number of handwriting points corresponding to the second preset interval; for example, taking the second coordinate axis as the Y axis, the Y axis coordinate of the projection histogram represents the position of the handwriting point in the first handwriting information in the Y direction, and the X axis coordinate represents the number of handwriting points mapped onto different Y axis positions; for example, the number of handwriting points in the projection histogram in the 0-100 interval of the Y axis is down-sampled to the new 0-10 interval, and the specific implementation principle is described above for the first projection histogram, which will not be repeated here.

[0187] In this embodiment, the projection histogram is used to represent the statistical diagram of the number of handwriting points of the first handwriting information projected on the preset coordinate axis.

[0188] Optionally, the generated table can be displayed in the display area, and the first handwriting information written by the user in the cells of the displayed table can be received.

[0189] In the embodiment, the transparency of the displayed table is adjustable, and the user can adjust the transparency of the table according to the requirement. In some embodiments, the displayed table can be completely transparent and the user cannot perceive that the user is writing in the table, thereby improving the user experience.

[0190] In some embodiments, the embodiment determines the second handwriting information bound to at least one cell from the first handwriting information according to the positional relationship between each cell in the generated table and the first handwriting information in any of the following ways:

[0191] Way 1) When the first handwriting information is located in the cell, the first handwriting information is determined as the second handwriting information bound to the cell.

[0192] In the implementation, if the first handwriting information is located in the cell, the first handwriting information is determined as the second handwriting information bound to the cell.

[0193] Way 2) When the landing point of the first handwriting information is located in the cell, the first handwriting information is determined as the second handwriting information bound to the cell.

[0194] In the implementation, the first handwriting information includes one or more strokes, and each stroke includes three actions of landing, writing, and lifting. Whether the stroke containing the landing point is bound to the cell can be determined according to the position of the landing point in the first handwriting information. For example, when the landing point in the first handwriting information is located in the cell, the stroke containing the landing point in the first handwriting information is determined as the second handwriting information bound to the cell.

[0195] Way 3) When the center of gravity of the first handwriting information is located in the cell, the first handwriting information is determined as the second handwriting information bound to the cell.

[0196] In the implementation, the first handwriting information includes one or more strokes, and each stroke has a center of gravity. Whether the stroke is bound to the cell can be determined according to the position of the center of gravity of the stroke in the first handwriting information. For example, when the center of gravity of the stroke in the first handwriting information is located in the cell, the stroke is determined as the second handwriting information bound to the cell.

[0197] Optionally, if the first handwriting information is partially located in the cell and partially exceeds the cell but does not exceed a set region centered on the cell, the first handwriting information is determined as the second handwriting information bound to the cell.

[0198] Optionally, if the part of the first handwriting information that is greater than the preset threshold is located in the cell or in a set region centered on the cell, the first handwriting information is determined as the second handwriting information bound to the cell.

[0199] In step 102, the second handwriting information bound to at least one cell is recognized to obtain the recognition data corresponding to the at least one cell.

[0200] In some embodiments, the second handwriting information bound to each cell can be recognized to obtain the recognition data corresponding to each cell. In an implementation, the recognition data corresponding to each cell is obtained by recognizing the second handwriting information corresponding to each cell.

[0201] In some embodiments, the size of the table in the present embodiment is adaptively adjusted according to the size of the second handwriting information bound to the cell. In an implementation, the size of the table is adaptively adjusted according to the size of the content written by the user, and it is to be noted that if the size of a cell is adjusted, the height of the row in which the cell is located and the width of the column in which the cell is located are also adjusted.

[0202] In some embodiments, when the second handwriting information bound to the cell exceeds the area covered by the cell, the row and the column in which the cell is located are expanded so that the second handwriting information is located in the area covered by the cell. In an implementation, the height of the row and the width of the column are appropriately increased so that the content written by the user in the cell is just located in the cell. As shown in FIG. 8, the present embodiment provides a schematic diagram of adaptive expansion of a table. When the content written by the user in the cell exceeds the coverage of the current cell, the height of the row and the width of the column in which the cell is located are expanded so that the handwriting information written by the user is located in the cell. Figure 2

[0203] Optionally, the area covered by the cell in the present embodiment can be the internal region of the cell. In the present embodiment, the size of the area covered by the cell can be the same as the size of the internal region of the cell, or can be slightly smaller than the size of the internal region of the cell. The size of the area covered by the cell is not limited in the present embodiment.

[0204] ​When the area covered by a cell is the internal area of the cell which is slightly smaller than the size of the cell, when the second handwriting information exceeds the area covered by the cell and does not exceed the boundary of the cell, the cell is expanded so that the content written by the user does not exceed the boundary of the cell.

[0205] In some embodiments, after the row and column where the cell is located are expanded when the second handwriting information bound by the cell exceeds the area covered by the cell, at least one offset cell in the table other than the expanded cell is determined, wherein the offset of the cell is based on the cell in the table after expansion; and the second handwriting information bound by the at least one offset cell is offset according to the offset amount of the at least one offset cell. As Figure 3 As shown in the figure, the present embodiment provides a schematic diagram of handwriting offset after table expansion. After the user writes "Zhang San" in a cell, the row height of the row where the cell is located and the column width of the column where the cell is located are expanded according to the size of "Zhang San". After expansion, there is a certain offset of the position of each cell relative to the position before expansion. According to the offset amount of the position of each cell, the second handwriting information bound by each cell is offset correspondingly, so that the offset second handwriting information can all fall in the center position or the preset position of the bound cell. The data recognition method provided by the present embodiment can ensure one-to-one correspondence between arrays, ensure that arrays in the same row and column are not misaligned, and thus ensure the correctness of array division, and further generate an accurate chart, no matter how the user writes, as long as the written content falls within the table.

[0206] In some embodiments, the size of the table in the present embodiment can also be fixed.

[0207] When the size of the table is fixed, the size of the content in each cell can be adaptively adjusted. In some embodiments, after obtaining the recognition data corresponding to the cell, the recognition data is converted into standard data conforming to a preset specification; and the standard data is scaled according to the size of the area covered by the cell, so that the standard data is located within the area covered by the cell.

[0208] As Figure 4 As shown in the figure, the present embodiment provides a schematic diagram of adaptive adjustment of content with fixed size of table. After the second handwriting information bound by any cell is recognized to obtain corresponding recognition data, the recognition data is converted into a character or a number in a format such as Songti, Kaishu, and / or Bold, and the character or the number is scaled according to the size of the cell, wherein the more the content written in the table, the smaller the font, so as to ensure that each content is located within the cell.

[0209] In some embodiments, the present embodiment is also used to perform at least two recognitions on the second handwriting information of the cell binding in the table.

[0210] In some embodiments, the at least two recognitions are performed by:

[0211] When it is determined that the second handwriting information of the cell binding has changed after the last recognition on the second handwriting information of the cell binding, the present recognition on the second handwriting information of the cell binding is performed.

[0212] In some embodiments, the determination that the second handwriting information of the cell binding has changed is performed by:

[0213] comparing the second handwriting information of the cell binding before the present recognition and the second handwriting information of the cell binding at the last recognition;

[0214] When the comparison result is consistent, the present recognition on the second handwriting information of the cell binding is skipped.

[0215] When the comparison result is inconsistent, it is determined that the second handwriting information of the cell binding has changed, and the present recognition on the second handwriting information of the cell binding is performed.

[0216] In some embodiments, when the comparison result is inconsistent, the present recognition on the second handwriting information of the cell binding in the table is performed by using a second recognition algorithm, and the last recognition is performed by using a first recognition algorithm. The first recognition algorithm and the second recognition algorithm are different.

[0217] In some embodiments, when the comparison result is inconsistent, the present recognition on the second handwriting information of the cell binding is performed by using a recognition algorithm different from that of the last recognition. In some embodiments, the first recognition algorithm includes a recurrent neural network algorithm, such as a recurrent neural network (RNN) or a long short term memory network (LSTM), and the second recognition algorithm includes a convolutional neural network algorithm, such as a convolutional neural network (CNN).

[0218] In some embodiments, the changed second handwriting information is obtained by at least one of modification, deletion, and addition on the second handwriting information of the cell binding after the last recognition on the second handwriting information of the cell binding.

[0219] In some embodiments, when the comparison result is inconsistent, the second handwriting information of the cell binding before the current identification is obtained by at least one of the following operations of the user on the second handwriting information of the cell binding in the last identification: modification, deletion, and addition.

[0220] In an implementation, when the user writes content in the table and performs identification, and then modifies, deletes, or adds (including adding before the original content or adding in the middle of the original content) the written content, the modified second handwriting information of the cell binding can be identified separately.

[0221] In some embodiments, when multiple cells in the table are bound to the second handwriting information, the second handwriting information of the multiple cells is identified in parallel; the correspondence between the identification data obtained by identification and the multiple cells is determined according to the row and column information of the cells bound to the second handwriting information, to obtain identification data corresponding to the multiple cells respectively.

[0222] In an implementation, after the user writes by hand in the table, each handwriting of the user is bound to the corresponding table, that is, each cell of the current table is bound to handwriting information, and each handwriting is bound to only one cell, and each cell stores a handwriting set bound in the current identification and a handwriting set bound in the last identification, where the handwriting set includes one or more second handwriting information.

[0223] In an implementation, after the user stops writing, a response time can be set, for example, 3 seconds, and after the user stops writing for 3 seconds, the second handwriting information bound to each cell in the entire table is submitted to a character recognition algorithm for identification, where the row and column information such as index values (row index value and column index value) of each cell are also submitted to the character recognition algorithm, so that the identification result returned by the algorithm has a correspondence with the index values, and the identification results of different cells are distinguished by the index values.

[0224] In some embodiments, after the identification data corresponding to the multiple cells is obtained, a chart can be generated according to the identification data corresponding to the multiple cells and the row and column information of each cell in the table, and the generated chart can be displayed.

[0225] Optionally, the chart type generated by the present embodiment includes at least one or more of the following: a column chart, a pie chart, a radar chart, a line chart, and a histogram.

[0226] As shown in FIG. 1, Figure 5 The present embodiment also provides a recognition method based on chart drawing data, and the implementation process of the method is as follows:

[0227] Step 501, entering a chart drawing interface, displaying a table in a display area;

[0228] Step 502, receiving first handwriting information written by a user in a cell of the displayed table;

[0229] Step 503, determining the first handwriting information as second handwriting information bound to the cell when the first handwriting information is located in the cell;

[0230] Step 504, expanding a row and a column where the cell is located when the second handwriting information bound to the cell exceeds an area covered by the cell, determining at least one offset cell in the table, and offsetting the second handwriting information bound to the at least one offset cell according to an offset amount of the at least one offset cell;

[0231] Step 505, identifying the second handwriting information bound to multiple cells in the table in parallel when the multiple cells in the table are bound to the second handwriting information, determining a correspondence between recognition data obtained by the identification and the multiple cells according to row and column information of the cells bound to the second handwriting information in the table, and obtaining recognition data corresponding to the multiple cells respectively;

[0232] Step 506, identifying the second handwriting information bound to a cell in the table at least twice, and comparing the second handwriting information bound to the cell before the present identification and the second handwriting information bound to the cell when the last identification is performed before the present identification is performed;

[0233] Step 507, determining whether the comparison result is consistent, and performing step 508 if the comparison result is consistent, or performing step 509 if the comparison result is not consistent;

[0234] Step 508, skipping the identification of the second handwriting information bound to the cell this time;

[0235] Step 509, identifying the second handwriting information bound to the cell this time.

[0236] As shown in the following table drawing data based on the identification method provided by the embodiment: Figure 6 The implementation process of the method is shown in the following table drawing data based on the identification method provided by the embodiment:

[0237] Step 601, entering a chart drawing interface, displaying a table in a display area;

[0238] Step 602, receiving first handwriting information written by a user in a cell of the displayed table;

[0239] Step 603, when the first handwriting information is determined as the second handwriting information bound to the cell when the landing point of the first handwriting information is located in the cell;

[0240] Step 604, when multiple cells in the table are bound to the second handwriting information, the second handwriting information bound to the multiple cells is identified in parallel, the correspondence between the recognition data obtained through identification and the multiple cells is determined according to the row and column information of the cells bound to the second handwriting information in the table, and the recognition data corresponding to the multiple cells is obtained respectively;

[0241] Step 605, the recognition data of each cell is converted into standard data conforming to a preset specification, and the standard data is scaled according to the size of the area covered by each cell;

[0242] Step 606, the second handwriting information bound to the cell in the table is identified at least twice, and before the second handwriting information bound to the cell is identified this time, the second handwriting information bound to the cell before this time and the second handwriting information bound to the cell when the last time is identified are compared;

[0243] Step 607, whether the comparison result is consistent is judged, if yes, step 608 is executed, otherwise step 609 is executed;

[0244] Step 608, the identification of the second handwriting information bound to the cell this time is skipped;

[0245] Step 609, the second handwriting information bound to the cell is identified this time.

[0246] Embodiment 2, based on the same inventive concept, the present application embodiment also provides an intelligent interactive display device, since the device is the device in the method in the present application embodiment, and the principle of solving the problem of the device is similar to that of the method, therefore, the implementation of the device can be referred to the implementation of the method, and the repeated parts will not be repeated.

[0247] As shown in Figure 7 , the device includes a display screen 700 and a control circuit 701, wherein:

[0248] The display screen 700 is used for displaying content;

[0249] The control circuit 701 includes a processor and a memory, the memory is used for storing programs executable by the processor, and the processor is used for reading programs in the memory and executing the following steps:

[0250] Receiving first handwriting information written by a user;

[0251] determine, from the first handwriting information, second handwriting information bound to at least one cell according to a positional relationship between each cell in the generated table and the first handwriting information;

[0252] recognize the second handwriting information bound to the at least one cell to obtain recognition data corresponding to the at least one cell.

[0253] As an optional implementation,

[0254] The table is generated according to the first handwriting information written by the user; or,

[0255] The table is generated according to preset row and column parameters.

[0256] As an optional implementation, the processor is specifically configured to perform:

[0257] display the generated table in a display area;

[0258] receive first handwriting information written by the user in a cell of the displayed table.

[0259] As an optional implementation, the processor is specifically configured to perform:

[0260] when the first handwriting information is located in the cell, determine the first handwriting information as second handwriting information bound to the cell; or,

[0261] when a landing point of the first handwriting information is located in the cell, determine the first handwriting information as second handwriting information bound to the cell; or,

[0262] when a center of gravity of the first handwriting information is located in the cell, determine the first handwriting information as second handwriting information bound to the cell.

[0263] As an optional implementation, a size of the table is adaptively adjusted according to a size of the second handwriting information bound to the cell.

[0264] As an optional implementation, the processor is specifically further configured to perform:

[0265] when second handwriting information bound to a cell exceeds an area covered by the cell, expand a row and a column where the cell is located, so that the second handwriting information is located in the area covered by the cell.

[0266] As an optional implementation, the processor is specifically further configured to perform:

[0267] determining offsets of at least one cell in the table except the expanded cell, wherein the offsets of the cells are based on the cells in the table after expansion;

[0268] offsetting second handwriting information bound to the at least one offset cell according to the offsets of the at least one offset cell.

[0269] As an optional implementation, after obtaining the recognition data corresponding to the cell, the processor is specifically further configured to perform:

[0270] convert the recognition data into standard data conforming to a preset specification;

[0271] scale the standard data according to a size of an area covered by the cell, so that the standard data is located in the area covered by the cell.

[0272] As an optional implementation, the processor is specifically further configured to perform:

[0273] perform at least two recognitions on the second handwriting information bound to the cells in the table.

[0274] As an optional implementation, the processor is specifically configured to perform:

[0275] when it is determined that the second handwriting information bound to the cell has changed after the last recognition on the second handwriting information bound to the cell, perform the recognition on the second handwriting information bound to the cell this time.

[0276] As an optional implementation, the processor is specifically configured to perform:

[0277] compare the second handwriting information bound to the cell this time before the recognition and the second handwriting information bound to the cell when last recognized;

[0278] when the comparison result is inconsistent, determine that the second handwriting information bound to the cell has changed.

[0279] As an optional implementation, the processor is specifically configured to perform:

[0280] when the comparison result is inconsistent, perform the recognition on the second handwriting information bound to the cells in the table this time by using a second recognition algorithm;

[0281] wherein the first recognition algorithm is used when last recognized.

[0282] As an optional implementation,

[0283] The first recognition algorithm comprises a recurrent neural network algorithm, and the second recognition algorithm comprises a convolutional neural network algorithm.

[0284] As an optional implementation,

[0285] The changed second handwriting information is obtained through at least one of modification, deletion and addition of the second handwriting information of the cell binding by the user after last recognition of the second handwriting information of the cell binding.

[0286] As an optional implementation, the processor is specifically configured to perform:

[0287] When multiple cells in the table are bound with second handwriting information, the second handwriting information of the multiple cells is recognized in parallel;

[0288] According to the row and column information of the cell bound with the second handwriting information in the table, a corresponding relationship between the recognition data obtained through recognition and the multiple cells is determined, and the recognition data corresponding to the multiple cells respectively is obtained.

[0289] As an optional implementation, the processor is specifically configured to perform:

[0290] After the recognition data corresponding to the multiple cells is obtained, a graph is generated according to the recognition data corresponding to the multiple cells and the row and column information of each cell in the table.

[0291] As an optional implementation, the transparency of the table display is adjustable.

[0292] As an optional implementation, the table is not displayed in the display area.

[0293] Embodiment 3: Based on the same inventive concept, the present application embodiment further provides an intelligent interactive display device. Since the device is the device in the method in the present application embodiment, and the principle of solving problems of the device is similar to that of the method, the implementation of the device can be referred to the implementation of the method, and the repeated parts will not be described here.

[0294] As Figure 8 shown, the device comprises:

[0295] The receiving unit 800 is configured to receive first handwriting information written by a user.

[0296] The binding unit 801 is configured to determine second handwriting information bound with at least one cell from the first handwriting information according to a position relationship between each cell in the generated table and the first handwriting information.

[0297] The recognition unit 802 is configured to recognize the second handwriting information of the at least one cell binding, and obtain the recognition data corresponding to the at least one cell.

[0298] As an optional implementation,

[0299] The table is generated according to the first handwriting information written by the user; or,

[0300] The table is generated according to preset row and column parameters.

[0301] As an optional implementation, the receiving unit 800 is specifically configured to:

[0302] display the generated table in a display area;

[0303] receive the first handwriting information written by the user in the cells of the displayed table.

[0304] As an optional implementation, the binding unit 801 is specifically configured to:

[0305] when the first handwriting information is located in the cell, determine the first handwriting information as the second handwriting information of the cell binding; or,

[0306] when the landing point of the first handwriting information is located in the cell, determine the first handwriting information as the second handwriting information of the cell binding; or,

[0307] when the center of gravity of the first handwriting information is located in the cell, determine the first handwriting information as the second handwriting information of the cell binding.

[0308] As an optional implementation, the size of the table is adaptively adjusted according to the size of the second handwriting information of the cell binding.

[0309] As an optional implementation, the method further includes an expansion unit specifically configured to:

[0310] when the second handwriting information of the cell binding exceeds the area covered by the cell, expand the row and column where the cell is located, so that the second handwriting information is located in the area covered by the cell.

[0311] As an optional implementation, the method further includes an offset unit specifically configured to:

[0312] determine the offset of at least one cell in the table except the expanded cell, wherein the offset of the cell is based on the cell expanded in the table;

[0313] According to the offset of the at least one offset cell, the second handwriting information bound to the at least one offset cell is offset correspondingly.

[0314] As an optional implementation, after the recognition data corresponding to the cell is obtained, the scaling unit is specifically configured to:

[0315] convert the recognition data into standard data conforming to a preset specification;

[0316] scale the standard data according to the size of the area covered by the cell, so that the standard data is located in the area covered by the cell.

[0317] As an optional implementation, the system further comprises a re-recognition unit specifically configured to:

[0318] recognize the second handwriting information bound to the cell in the table at least twice.

[0319] As an optional implementation, the re-recognition unit is specifically configured to:

[0320] when it is determined that the second handwriting information bound to the cell has changed after the last recognition of the second handwriting information bound to the cell, recognize the second handwriting information bound to the cell this time.

[0321] As an optional implementation, the re-recognition unit is specifically configured to:

[0322] compare the second handwriting information bound to the cell this time before the current recognition with the second handwriting information bound to the cell at the last recognition;

[0323] when the comparison result is inconsistent, it is determined that the second handwriting information bound to the cell has changed.

[0324] As an optional implementation, the re-recognition unit is specifically configured to:

[0325] when the comparison result is inconsistent, use a second recognition algorithm to recognize the second handwriting information bound to the cell in the table this time;

[0326] wherein the first recognition algorithm is used at the last recognition.

[0327] As an optional implementation,

[0328] the first recognition algorithm comprises a recurrent neural network algorithm, and the second recognition algorithm comprises a convolutional neural network algorithm.

[0329] As an optional implementation,

[0330] The changed second handwriting information is obtained by at least one of modifying, deleting, or adding the second handwriting information of the cell binding after the user recognizes the second handwriting information of the cell binding last time.

[0331] As an optional implementation, the identification unit 802 is specifically configured to:

[0332] When multiple cells in the table are bound with the second handwriting information, the second handwriting information bound with the multiple cells is identified in parallel.

[0333] According to the row and column information of the cell bound with the second handwriting information in the table, a corresponding relationship between the identification data obtained by identification and the multiple cells is determined, and identification data corresponding to the multiple cells is obtained.

[0334] As an optional implementation, the system further includes a chart generation unit specifically configured to:

[0335] After the identification data corresponding to the multiple cells is obtained, a chart is generated according to the identification data corresponding to the multiple cells and the row and column information of each cell in the table.

[0336] As an optional implementation, the transparency of the table display is adjustable.

[0337] As an optional implementation, the table is not displayed in the display area.

[0338] Based on the same inventive concept, the embodiments of the present application further provide a computer storage medium, which stores a computer program, and the program is executed by a processor to implement the following steps:

[0339] Receiving first handwriting information written by a user;

[0340] According to the position relationship between each cell in the generated table and the first handwriting information, determining second handwriting information bound with at least one cell from the first handwriting information;

[0341] Identifying the second handwriting information bound with the at least one cell to obtain identification data corresponding to the at least one cell.

[0342] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer usable program code.

[0343] The present application is described in reference to the drawings using a flowchart and / or a block diagram of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagram, and combinations of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 apparatus that performs the functions specified in the flowchart and / or block diagram block or blocks.

[0344] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart and / or block diagram block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 apparatus that performs the functions specified in the flowchart and / or block diagram block or blocks.

[0345] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that execute on the computer or other programmable apparatus provide steps for implementing the flowchart and / or block diagram block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 apparatus that performs the functions specified in the flowchart and / or block diagram block or blocks.

[0346] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the apparatus and methods disclosed herein, the application can be practiced otherwise than as specifically described.

Claims

1. A method of data recognition, characterized by, The method is applied to an intelligent interactive display device, and comprises the following steps: receiving first handwriting information written by a user; determining second handwriting information bound to at least one cell from the first handwriting information according to a positional relationship between each cell in a generated table and the first handwriting information; recognizing the second handwriting information bound to at least one cell to obtain recognition data corresponding to the at least one cell; when the size of the table is fixed, the size of the recognition data in each cell is adaptively adjusted; when the recognition data corresponding to multiple cells is obtained, generating a chart according to the recognition data corresponding to the multiple cells and row and column information of each cell in the table; each cell in the table is generated by any one or more of the following ways: way 1) determining column separation lines according to valley positions of the first handwriting information in a first projection histogram of a first coordinate axis; determining row separation lines of any one column according to valley positions of the first handwriting information contained in the column in a second projection histogram of a second coordinate axis; and determining each cell according to the column separation lines and the row separation lines; way 2) determining row separation lines according to valley positions of the first handwriting information in the second projection histogram of the second coordinate axis; determining column separation lines of any one row according to valley positions of the first handwriting information contained in the row in the first projection histogram of the first coordinate axis; and generating each cell of the table according to the row separation lines and the column separation lines; way 3) determining column separation lines according to valley positions of the first handwriting information in the first projection histogram of the first coordinate axis; determining row separation lines according to valley positions of the first handwriting information in the second projection histogram of the second coordinate axis; and determining each cell according to the row separation lines and the column separation lines.

2. The method of claim 1, wherein: the table is generated according to the first handwriting information written by the user; or the table is generated according to preset row and column parameters.

3. The method of claim 1, wherein, The receiving of the first handwriting information written by the user comprises: displaying the generated table in a display area; receiving the first handwriting information written by the user in a cell of the displayed table.

4. The method of claim 1, wherein, The determination of the second handwriting information bound to at least one cell from the first handwriting information according to a positional relationship between each cell in a generated table and the first handwriting information comprises: when the first handwriting information is located in the cell, determining the first handwriting information as the second handwriting information bound to the cell; or when a landing point of the first handwriting information is located in the cell, determining the first handwriting information as the second handwriting information bound to the cell; or when a center of gravity of the first handwriting information is located in the cell, determining the first handwriting information as the second handwriting information bound to the cell.

5. The method according to any one of claims 1 to 4, characterized in that, The size of the table is adaptively adjusted according to the size of the second handwriting information bound to the cell.

6. The method of claim 5, wherein, Further comprising: when the second handwriting information bound to the cell exceeds an area covered by the cell, expanding a row and a column where the cell is located so that the second handwriting information is located in the area covered by the cell.

7. The method of claim 6, wherein, Further comprising: determining at least one cell offset in the table except the expanded cell, wherein the offset of the cell is based on the cell in the table after expansion; according to the offset of the at least one offset cell, offsetting the second handwriting information bound to the at least one offset cell.

8. The method of any one of claims 1-4, wherein, After obtaining the recognition data corresponding to the cell, further comprising: converting the recognition data into standard data conforming to a preset specification; scaling the standard data according to the size of the area covered by the cell, so that the standard data is located within the area covered by the cell.

9. The method of any one of claims 1-4, wherein, Further comprising: performing at least two recognitions on the second handwriting information bound to the cell in the table.

10. The method of claim 9, wherein, The at least two recognitions on the second handwriting information bound to the cell in the table, comprising: when determining that the second handwriting information bound to the cell has changed after the last recognition on the second handwriting information bound to the cell, performing the current recognition on the second handwriting information bound to the cell.

11. The method of claim 10, wherein, The determination that the second handwriting information bound to the cell has changed, comprising: comparing the second handwriting information bound to the cell before the current recognition with the second handwriting information bound to the cell in the last recognition; when the comparison result is inconsistent, determining that the second handwriting information bound to the cell has changed.

12. The method of claim 11, wherein, When the comparison result is inconsistent, the current recognition on the second handwriting information bound to the cell, comprising: when the comparison result is inconsistent, using a second recognition algorithm to perform the current recognition on the second handwriting information bound to the cell in the table; wherein the first recognition algorithm is used in the last recognition.

13. The method of claim 12, wherein: the first recognition algorithm comprises a recurrent neural network algorithm; and the second recognition algorithm comprises a convolutional neural network algorithm.

14. The method of claim 10, wherein: the changed second handwriting information is obtained by at least one of modification, deletion, and addition on the second handwriting information bound to the cell after the last recognition on the second handwriting information bound to the cell.

15. The method of any one of claims 1-4, wherein, The recognition on the second handwriting information bound to at least one cell to obtain the recognition data corresponding to the at least one cell, comprising: when multiple cells in the table are bound to the second handwriting information, performing the recognition on the second handwriting information bound to the multiple cells in parallel; determining the correspondence between the recognition data obtained by recognition and the multiple cells according to the row and column information of the cells bound to the second handwriting information in the table, to obtain the recognition data corresponding to the multiple cells respectively.

16. The method of any one of claims 1-4, wherein, The transparency of the table display is adjustable.

17. The method of claim 1 or 2, wherein, The table does not display in the display area.

18. An intelligent interactive display device, characterized by It comprises a display screen and a control circuit, wherein: the display screen is used for displaying content; the control circuit comprises a processor and a memory, the memory is used to store programs executable by the processor, and the processor is used to read the programs in the memory and execute the steps of the method of any one of claims 1-17.

19. A computer storage medium having stored thereon a computer program, characterized in that The program, when executed by the processor, implements the steps of the method as claimed in any one of claims 1 to 17.

Images