A black-and-white printed paper piece code point recognition method for an intelligent pen and paper

Abstract

The application discloses a black-and-white printed paper code point identification method for an intelligent pen and paper, which comprises the following steps: pre-processing homework content to obtain homework printed files containing code point information; obtaining paper containing code point information through black-and-white printing of the homework printed files; obtaining handwriting data by recording code point information and handwriting in a specified area of the paper in real time through an intelligent pen; decoding and correcting the code points on the handwriting data through the intelligent pen to obtain local handwriting images, and finally transmitting the local handwriting images to a server; and the server collects each local handwriting image according to the position relationship corresponding to the code points to obtain overall handwriting content. The homework can be printed by using a common black-and-white printer, the printing cost is lower, the printing speed is faster, and the requirement for the equipment is lower; meanwhile, the homework is not easily affected by answer lines, local code point damage or unclear printing, and the automatic error correction capability is strong; meanwhile, the application can effectively monitor handwriting state and pen holding state, and monitor whether the pen is broken.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent pen, in particular to a black and white printed paper code point identification method for an intelligent pen. BACKGROUND

[0002] The existing intelligent pen can usually only identify color printing or high-precision printed dot matrix paper. However, with the popularization of printers, many families also choose to print dot matrix paper by themselves to cooperate with the intelligent pen for homework. However, due to factors such as the model and precision of small household printers, partial code points may be contaminated or unclear, resulting in abnormal recognition; especially when black and white printing is encountered with continuous interference lines, the error rate of recognition will be further increased. The existing solution is usually to use color inkjet printing, and code points are printed with black ink containing carbon, and texts are mixed with red, blue and yellow; this way is slow in printing speed, and more color ink is needed, which is high in use cost. In order to improve the application range of the intelligent pen and make it adapt to more ordinary household small black and white printers, reduce the use cost and improve the printing speed, there is an urgent need for an identification method that can efficiently and accurately read code points on black and white printed paper. SUMMARY

[0003] In view of the above problems, the present application aims to provide a black and white printed paper code point identification method for an intelligent pen.

[0004] To achieve the technical purpose, the scheme of the present application is: a black and white printed paper code point identification method for an intelligent pen, the homework content is preprocessed to obtain a homework printed file containing code point information; the homework printed file is printed by black and white printing to obtain a paper containing code point information;

[0005] The code point information and writing traces of the specified area on the paper are recorded in real time by the intelligent pen to obtain trace data; the code points on the trace data are decoded and corrected by the intelligent pen to obtain a local writing image, which is finally transmitted to a server; the server collects each local writing image according to the position relationship of the code points to obtain the overall writing content.

[0006] As a preferred, when recording, the single shooting area of the intelligent pen is capable of covering 4-9 complete code points; when decoding, the code point A0 ij printed in the code area where the code point A0 ij is located is analyzed to obtain the corresponding coordinate information (i, j);

[0007] If the code point A0 is contaminated, damaged or partially covered by other text lines, the code point A0 cannot be analyzed, then other code points in the local writing image are analyzed until the code point A0 mn capable of obtaining the coordinate information is obtained after moving the coordinates (a, b), and the code point A0 mnThe coordinates of the code point A0 are (m-a, n-b) corresponding to the coordinate information (m, n);

[0008] During correction and restoration, the image of the handwriting data is adjusted by distortion to obtain a trapezoidal image, and the trapezoidal image is corrected and restored to obtain a local writing image.

[0009] Preferably, the size of the code point is 1.5 mm*1.5 mm, and the resolution of the smart pen is greater than or equal to 240*240.

[0010] Preferably, the smart pen also records the writing pressure data of the writing process in real time while shooting; when the writing pressure value is less than or equal to the initial value, it is judged that the writing has not started or the writing is finished in the pen-up state, and the smart pen stops shooting; when the writing pressure value is greater than the initial value, the writing action is being performed, and the smart pen keeps shooting continuously.

[0011] The smart pen also records the pen-holding data of the writing process in real time while shooting, and the pen-holding data includes the pen-holding direction and the pen-holding angle. When the pen-holding direction or the pen-holding angle does not meet the specifications, a reminder is given. The pen-holding angle data is also used to assist the correction and restoration of the handwriting data.

[0012] Preferably, the local writing image is marked with the coordinate information corresponding to the code point, and then the handwriting on the adjacent local writing images is spliced and combined to obtain the overall writing content with complete handwriting.

[0013] When the writing pressure value is greater than the initial value, there is no handwriting in the code area of the code point, and there is handwriting in the code area of the adjacent code point, there is a partial broken pen, and the endpoints of the handwriting in the code area of the adjacent code point are connected and combined. The writing direction and stroke characteristics are used to select the connected line that meets the requirements.

[0014] When the writing pressure value is greater than the initial value, there is no handwriting in the code area of the code point, and there is no handwriting in the code area of the adjacent code point, there is a continuous broken pen, and if the number of continuous broken pens exceeds a threshold value, a reminder is given.

[0015] Preferably, when the single shooting area of the smart pen can cover 4 complete code points, the coordinate information of each code point on each local writing image is obtained, the overlapping code points on the two local writing images adjacent to the code point change are analyzed, and the direction position of the overlapping code points on the first local writing image is the writing direction.

[0016] When the single shooting area of the smart pen can cover 9 complete code points, the coordinate information of each code point on each local writing image is obtained to determine the center code point, and the connected line of the center code points on the local writing images adjacent to the code point change is the writing direction.

[0017] As preferred, the code point is a square, a diagonal line is arranged on the code point, the diagonal line divides the code point into a first region and a second region, and the diagonal line of the code point is perpendicular to the diagonal lines of the code points in the four directions above, below, left and right thereof; four adjacent first regions constitute a first position reference square, and four adjacent second regions constitute a second position reference square.

[0018] When the single shooting region of the intelligent pen can completely cover the first position reference square or the second position reference square, the coordinate information of the first region and the second region on each code point on each local writing image is acquired.

[0019] An intelligent pen applying the code point recognition method for the black-and-white printed paper of the intelligent pen, comprising:

[0020] A sensor is arranged for collecting and acquiring handwriting data, writing pressure data and pen holding data generated by a user during writing on paper.

[0021] A Bluetooth communication module is arranged for wirelessly communicating with a pen box and transmitting the handwriting data to the pen box.

[0022] A power module is arranged for providing power for the sensor, the Bluetooth communication module and the processor in the intelligent pen.

[0023] A processor is arranged for analyzing and processing the handwriting data, the writing pressure data and the pen holding data.

[0024] A computer readable storage medium is arranged for storing computer instructions for causing the computer to execute the code point recognition method for the black-and-white printed paper of the intelligent pen.

[0025] The application has the advantages that the work of the application can be printed by using a common black-and-white printer, the printing cost is lower, the printing speed is faster, and the requirement for the equipment is lower; meanwhile, the application is not easily affected by the answer line, the local code point damage or unclear printing, and has strong automatic error correction capability; meanwhile, the application can effectively monitor the writing state and the pen holding state, and monitor whether the pen is broken. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 The flowchart of the first embodiment of the application;

[0027] Figure 2 The flowchart of the second embodiment of the application;

[0028] Figure 3 The flowchart of the third embodiment of the application;

[0029] Figure 4 The principle diagram of the intelligent pen of the application. DETAILED DESCRIPTION

[0030] The application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Figures 1-4 The application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Embodiments

[0031] A black-and-white printed paper point code recognition method for a smart pen, the specific steps are as follows:

[0032] S101, preprocessing the work content to obtain a work print file containing code point information;

[0033] S102, obtaining a paper containing code point information by black-and-white printing the work print file; the size of the code point is 1.5mm*1.5mm;

[0034] S103, obtaining handwriting data by real-time recording the code point information and handwriting of the specified area of the paper through the camera module of the smart pen, the resolution of the smart pen is greater than or equal to 240*240; when recording, the single shooting area of the smart pen can cover 4-9 complete code points; after decoding and correcting the code points on the handwriting data, a local handwriting image is obtained, and finally transmitted to the server; when decoding, the code point A0 printed in the code area is analyzed to obtain the corresponding coordinate information (i, j); if the code point is contaminated, damaged, or partially covered by other text lines, the code point A0 cannot be analyzed, then other code points in the local handwriting image are analyzed until the coordinate (a, b) is moved to obtain a code point A0 that can be analyzed to obtain the corresponding coordinate information (m, n), and the coordinate of the code point A0 is (m-a, n-b); ij ij mn mn

[0035] When correcting and restoring, the image of the handwriting data is obtained by distortion adjustment to obtain a trapezoidal image, and the trapezoidal image is corrected and restored to obtain a local handwriting image.

[0036] S104, the server collects each local handwriting image to obtain the overall handwriting content according to the position relationship of the code points.

[0037] The method of the application can effectively solve the interference of the pure black underline on the code points (the traditional method needs to print the code points with black, and the other printed content needs to be printed with three primary colors, which is high in cost). Normal Chinese grid and underlined test papers can be normally used. Since the resolution of the camera module is higher, the computing power required for correction and restoration is higher, and generally 240*240 resolution is used, and a photo can cover a 2*2 or 3*3 area. Embodiments

[0038] ​​​​A black and white printed paper code point recognition method for a smart pen, in order to better recognize the pen start, pen stop or pen lifting in the writing action, and the writing pressure in the continuous writing; therefore, a pressure sensor is used to collect the writing pressure value (a precise pressure collection module is formed by using a magnet and a pressure sensor that attract each other), and the pressure value collected by the pressure sensor when not writing is the initial value; at the same time, a gyroscope is used to collect parameters such as the angle, direction and angular velocity of holding the pen.

[0039] S201, while the smart pen is recording, the writing pressure data of the writing process is also recorded in real time;

[0040] S202, when the writing pressure value is less than or equal to the initial value, it is judged that the writing has not started or the writing is finished and the pen is lifted, and the smart pen stops recording; when the writing pressure value is greater than the initial value, the writing action is being performed, and the smart pen continues to record;

[0041] S203, while the smart pen is recording, the holding pen data of the writing process is also recorded in real time, and the holding pen data includes the holding pen direction and the holding pen angle;

[0042] S204, when the holding pen direction or the holding pen angle does not meet the specifications, a reminder is issued; at the same time, the holding pen angle data is used to assist the correction and restoration of the handwriting data. Embodiment

[0043] A black and white printed paper code point recognition method for a smart pen, in order to avoid or reduce the broken pen situation in the collection, the broken pen needs to be self-repaired or warned. At the same time, since the number of code points collected in each local writing image of the present application is greater than 1, the writing direction needs to be judged by comparing the overlapping code points, which is different from the traditional low-resolution direction judgment method.

[0044] S301, the local writing image is marked by the coordinate information corresponding to the code point, and then the handwriting on the adjacent local writing image is spliced and combined, and the overall writing content with complete handwriting is obtained by summarizing;

[0045] S302, when the writing pressure value is greater than the initial value, but there is no handwriting in the code area of the code point, and there is handwriting in the code area of the adjacent code point, there is a local broken pen, the end points of the code area with handwriting of the adjacent code point are connected and combined, and the connecting line meeting the requirements is selected by the writing direction and the stroke characteristics;

[0046] S303, when the single shot area of the smart pen can cover 4 complete code points, the coordinate information of each code point on each local writing image is obtained, the overlapping code points on the two local writing images adjacent to each other and having code point changes are analyzed, and then the direction position of the overlapping code points on the first local writing image is the writing direction; when the single shot area of the smart pen can cover 9 complete code points, the coordinate information of each code point on each local writing image is obtained to determine the center code point, and the line connecting the center code points on the local writing images adjacent to each other and having code point changes is the writing direction.

[0047] S304, when the writing pressure value is greater than the initial value, but there is no handwriting in the code area of the code point, and there is also no handwriting in the code area of the adjacent code point, there is a continuous pen break, and if the number of code points of the continuous pen break exceeds the threshold value, a prompt is sent. Embodiment

[0048] A black and white printed paper code point recognition method for a smart pen, in order to because the resolution of the existing smart pen is usually less than 180*180, and the processing capability is also weak, and the installation difficulty of a larger size photographic module (CCD) is also larger. Only a little more computing power is needed to deal with the interference of the oblique line on the code point. The code point is a square, and a diagonal line (a dotted line composed of a plurality of points) is arranged on the code point, so as to divide the code point into a first area and a second area, and the diagonal line of the code point is perpendicular to the diagonal lines of the code points in the four directions above, below, left and right thereof. Four adjacent first areas form a first position reference square, and four adjacent second areas form a second position reference square. When the single shot area of the smart pen can completely cover the first position reference square or the second position reference square, the coordinate information of the first area and the second area on each code point on each local writing image is obtained.

[0049] A smart pen applying the black and white printed paper code point recognition method for a smart pen, comprising:

[0050] A sensor for collecting handwriting data, writing pressure data and pen holding data generated by a user during writing on paper;

[0051] A Bluetooth communication module capable of wirelessly communicating with a pen box and transmitting the handwriting data to the pen box;

[0052] A power module for providing power for the sensor, the Bluetooth communication module and the processor in the smart pen;

[0053] A processor capable of analyzing and processing the handwriting data, the writing pressure data and the pen holding data.

[0054] A computer readable storage medium storing computer instructions for causing the computer to execute the aforementioned black and white printed paper code point recognition method for a smart pen.

[0055] The above description is only the preferred embodiment of the present application, and is not intended to limit the present application. Any slight modification, equivalent replacement and improvement made according to the technical essence of the present application to the above embodiment shall be included in the protection scope of the technical scheme of the present application.

Claims

1. A black and white printed paper piece code point recognition method for a smart pen, characterized in that: The work content is preprocessed to obtain a work print file containing code point information; The work print file is printed in black and white to obtain a paper document containing code point information; The code point information and writing traces of a specified area on the paper document are recorded in real time by the intelligent pen to obtain trace data; the code points on the trace data are decoded and corrected to obtain a local writing image, which is finally transmitted to a server; the server collects each local writing image according to the position relationship corresponding to the code points to obtain the overall writing content; When recording, the single shooting area of the intelligent pen can cover 4-9 complete code points; when decoding, the code point A ij printed in the code area where the code point A ij corresponding coordinate information (i, j) is analyzed to obtain If the code point is contaminated and damaged, or is partially covered by other text lines, and code point A0 cannot be parsed, then other code points in the partial writing image are parsed until the coordinate (a, b) is moved to obtain the code point A that can be parsed mn , and the corresponding coordinate information (m, n) of the code point A is obtained, then the coordinate of the code point A0 is (m-a, n-b) mn ; During correction and restoration, the image of the trace data is adjusted by distortion to obtain a trapezoidal image, and the trapezoidal image is corrected and restored to obtain a local writing image.

2. The black and white printed paper piece code point recognition method for the intelligent paper pen according to claim 1, characterized in that: The size of the code point is 1.5 mm*1.5 mm, and the resolution of the intelligent pen is greater than or equal to 240*240.

3. The black and white printed paper piece code point recognition method for the intelligent paper pen according to claim 2, characterized in that: The writing pressure data of the writing process is recorded in real time while the intelligent pen is recording; When the writing pressure value is less than or equal to the initial value, it is determined that the writing has not started or the writing is complete and the pen is lifted, and the intelligent pen stops recording; when the writing pressure value is greater than the initial value, the writing action is being performed, and the intelligent pen continues to record. The holding pen data of the writing process is recorded in real time while the intelligent pen is recording, and the holding pen data includes the holding pen direction and the holding pen angle; when the holding pen direction or the holding pen angle does not meet the specifications, a reminder is issued; and the holding pen angle data is used to assist the correction and restoration of the trace data.

4. The black and white printed paper piece code point recognition method for the intelligent paper pen according to claim 3, characterized in that: The local writing images are positionally marked by the coordinate information corresponding to the code points, and then the traces on adjacent local writing images are spliced and combined to obtain the overall writing content with complete traces; When the writing pressure value is greater than the initial value, but there is no trace in the code area of the code point, and there is a trace in the code area of the adjacent code point, there is a local pen break, and the endpoints of the code area of the adjacent code point with the trace are connected and combined, and the writing direction and stroke characteristics are used to select the connected line that meets the requirements; When the writing pressure value is greater than the initial value, but there is no trace in the code area of the code point, and there is no trace in the code area of the adjacent code point, there is a continuous pen break, and if the number of code points of the continuous pen break exceeds a threshold value, a reminder is issued.

5. The black and white printed paper piece code point recognition method for the intelligent paper pen according to claim 4, characterized in that: When the single shooting area of the intelligent pen can cover 4 complete code points, the coordinate information of each code point on each local writing image is obtained, the overlapping code points on two adjacent local writing images with code point changes are analyzed, and the direction position of the overlapping code points on the first local writing image is the writing direction; When the single shooting area of the intelligent pen can cover 9 complete code points, the coordinate information of each code point on each local writing image is obtained to determine the center code point, and the connected line of the center code points on the adjacent local writing images with code point changes is the writing direction.

6. The black and white printed paper piece code point recognition method for the intelligent paper pen according to claim 4, characterized in that: The code point is a square, and a diagonal line is arranged on the code point to divide the code point into a first area and a second area, and the diagonal line of the code point is perpendicular to the diagonal lines of the four adjacent code points; four adjacent first areas form a first position reference square, and four adjacent second areas form a second position reference square. When the single shot area of the smart pen can completely cover the first position reference block or the second position reference block, the coordinate information of the first area and the second area on each code point on each local writing image is acquired.

7. A smart pen applying the black-and-white printed paper code point recognition method for a smart pen according to any one of claims 1-6, characterized in that, The method comprises the steps of: a sensor for collecting handwriting data, writing pressure data and pen holding data generated by a user during writing on paper; a Bluetooth communication module capable of wirelessly communicating with a pen box and transmitting the handwriting data to the pen box; a power module for providing power for the sensor, the Bluetooth communication module and the processor in the smart pen; a processor capable of analyzing and processing the handwriting data, the writing pressure data and the pen holding data. 8.A computer readable storage medium storing computer instructions for causing the computer to execute the black and white printing paper code point identification method for a smart pen according to any one of the preceding claims 1-6.

Images