CN100472419C - handwritten analysis method - Google Patents

Abstract

The invention discloses a handwriting analysis method, which adopts the following steps for analysis: the subject uses a special pen for the input board to write on the surface of the input board connected to the computer, thereby obtaining the input writing trace; detecting the validity of the writing trace, Filter out invalid writing traces and keep valid writing traces; obtain various characteristic data of each trace point of valid writing traces and record them in the program memory; calculate the average value of some characteristic data, and automatically select multiple distributions according to the law The characteristic data of the track point, build a data sequence reflecting the change trend; display the specific values of various characteristic data, and give the change curve of various characteristic data; according to the specific value and change curve of the characteristic data, the writing track and process for analysis. It can not only accurately calculate the writing index of the testee and reflect the behavior characteristics of the writing process, but also image and digitally store the writing information, which is convenient for long-term preservation and other analysis.

Description

handwritten analysis method

technical field

The invention belongs to the handwriting analysis technology, and in particular relates to a handwriting analysis method which utilizes a computer program to analyze the writing track and process.

Background technique

So far, the existing handwriting analysis technology (mainly handwriting analysis technology) is based on the observation and analysis of the written shape and font (writing track), and a large amount of effective information such as writing speed and strength can only be obtained by the main tester. Speculation, the specific performance is: more phenomenological analysis, less empirical analysis; more holistic analysis (analysis of words and paragraphs), less basic analysis (analysis of basic writing strokes); more static analysis, less on the writing process Perform real-time dynamic tracking analysis. These characteristics determine that the previous analysis techniques are highly subjective and arbitrarily influenced by experience, which greatly reduces the reliability and validity of the analysis results.

Contents of the invention

In view of the shortcomings of the prior art, the purpose of the present invention is to provide a handwriting analysis method, which can not only accurately calculate the writing index of the subject, reflect the behavior characteristics of the writing process, but also image and digitally store the writing information, It is convenient for long-term storage and other analysis.

In order to achieve the above object, the technical solution of the present invention is: a handwriting analysis method, which utilizes a computer program to realize handwriting analysis, and it comprises the following steps:

(1) The subject uses the special pen for the input board to write on the surface of the input board connected to the computer, so as to obtain the input writing track;

(2) Detect the validity of the writing track, filter out the invalid writing track, and keep the valid writing track;

(3) obtain various feature data of each track point of effective writing track and record in the program memory, described feature data is the coordinate position, time point, distance, velocity value, acceleration value and pressure value of track point;

(4) Calculate the average value of some characteristic data, and automatically select the characteristic data of multiple distribution track points according to the law, set up a data sequence reflecting the changing trend, and calculate the average speed, average acceleration, and average speed of the entire writing track according to the data sequence. Pressure value, fluctuation value and breakpoint number;

(5) Display the specific values of various characteristic data, and give the change curve diagram of various characteristic data;

(6) Analyze the writing track and process according to the specific values and change curves of the characteristic data.

The obtained data sequence is saved in the storage database, and the records between different testers or the same tester before and after different tests can be called out as needed, and put together for visual comparison.

When the pressure value of a trajectory point is lower than 100, and the values of the two points before and after it are both higher than 100, it will be regarded as a fluctuation, and the fluctuation value will be increased by 1.

Based on the sum of the distances in the data sequence and the elapsed time from start to finish, the average speed of the entire trajectory is calculated.

According to the time and speed values at the beginning and end of the data sequence, the average acceleration of the entire trajectory from the beginning to the end is calculated.

According to the pressure values in the data series, the average writing pressure value of the entire trajectory is calculated.

The present invention belongs to different test modes from the existing handwriting (handwriting) analysis technology, and has the advantage of adopting a computerized test process to directly obtain the basic physical indicators of the writing process in the writing process, and the test data is objective and true, and can be extracted simply and conveniently Writing speed, acceleration, fluctuation index, and writing strength index, which cannot be extracted by traditional analysis techniques, have the advantages of strong objectivity of test results, high accuracy, and flexible test methods, which greatly improve the reliability of handwriting analysis and validity.

Description of drawings

The present invention will be further described in detail below in conjunction with the accompanying drawings.

Fig. 1 is a basic architecture diagram of the computing system of the method of the present invention.

Fig. 2 is a flowchart of the analysis operation of the method of the present invention.

Detailed ways

The present invention adopts the mode that computer software is combined with input board hardware (referring to Fig. 1, this figure only shows relevant parts relevant to the present invention, other frameworks such as input device keyboard, mouse etc., are not shown therein). The testee uses the input board as a writing input tool, and the system software judges and calculates synchronously during the process of receiving the data, calculates the speed, acceleration, writing fluctuation index and writing strength, stores the data in the database, and the main tester queries through the data Interface to view the quantitative test data of the subject and display it in a chart. The handwriting input part includes an input board and an input pen, which are used for writing by the tested person. Computing processor B is used to execute programs in memory. The system displays the processing information of the processing process on a display unit such as a liquid crystal display (LCD), providing the user with information related to the writing process. Since the computing processor and display components are common existing computer components, they will not be described in detail. The main program of the system includes main program modules such as main management program, writing feature calculation, storage database, and result analysis, etc., which are used to control the entire writing test process and complete data processing settings.

Using a computer program to realize handwriting analysis includes the following steps (see Figure 2):

First proceed to step 1, the main tester sets the test content and the information of the testee, the testee enters the data acquisition module of the application program, the computer screen displays the writing content and writing requirements that the testee needs to complete, and then enters step 2 ;

In step 2, the subject uses the special pen for the input board to write on the surface of the input board, obtains the input handwriting track, and then enters step 3;

In step 3, the program detects the validity of writing, filters out invalid traces, and only keeps valid writing parts. Then go to step 4;

In step 4, the calculation work is mainly completed. The program records the time point of the current position, the plane coordinates of the pen-down point and the strength of the writing pen pressing the input board, and calculates the plane distance from the previous position to the current coordinate position through physical formulas, The plane fluctuation range between two positions, the dynamic speed value, the acceleration value, these basic data are temporarily recorded in the program memory, in this step, the written graphics composed of the motion trajectory will be displayed on the screen interface at the same time, The speed, acceleration, and strength indicators here are a feature of this program, and the previous handwriting (handwriting) analysis cannot accurately calculate the values of these indicators. Then go to step 5;

In step 5, calculate based on the basic data temporarily recorded in the program memory, extract the overall average velocity, average acceleration, and average pressure, and automatically select the characteristic values of 24 distribution points according to the law, and build a reflection Serial data with a changing trend. When the user inputs the operation command "writing complete", go to step 6;

In step 6, save the target data in the storage database, complete the work of the data collection link, and enter step 7;

In step 7, enter the query and analysis interface of test results. In this step, the program displays the specific values of the characteristic data, and gives the change curves of various writing characteristics, such as the change graph of writing strength, and can put records between different testers and different tests before and after the same tester. Intuitive comparison together saves a lot of manpower and improves efficiency.

The following is a detailed introduction to the collection and sampling of feature data during the writing process:

Speed (velocity) In the program, the speed index indicates the speed of the testee's writing.

Distance between two points in 2D space: $p_1{p}_2=\sqrt{{(x_2-x_1)}^2+{({\mathrm{the\; y}}_2-{\mathrm{the\; y}}_1)}^2}---\left(1\right)$

Speed: v=distance (s)/time (t) (2)

Acceleration is used to measure the change of writing speed. The faster the acceleration, the greater the acceleration. If the speed remains unchanged, the acceleration is 0. If the object is decelerating, the acceleration is negative.

Acceleration: $a=\frac{\mathrm{\Δv}}{\mathrm{\Δt}}=\frac{v_f-v_i}{t_f-t_i}---\left(3\right)$

Strength (pressure) is used to represent the strength situation that the pen exerts on the input board when the subject uses the input pen to write on the input board. In the current program, the strength results are divided into 1024 horizontal grade representations.

Fluctuation is used to represent the frequency of fluctuations in the dynamics of the subject's writing process.

Each handwriting is a locus graph composed of at least 2 dot connections. When the subject starts to write on the surface of the input board with the stylus, the system clock of the program records the current time T, position coordinates X, Y, and divides the force into 1024 according to the pressure exerted by the stylus on the surface of the input board Horizontal grade, get the strength grade value Pressure of the previous position point. Because this is the first point of the entire trajectory, the initial velocity value is 0, the acceleration value is 0, and the distance is 0, and these index values are stored in the memory sequence according to the rules shown below.

When the subject continues to move the input pen to write (the pen moves to a new position), the program receives the data of the new position point, and calculates the speed according to the coordinate position and time of the two track points before and after.

Substitute the coordinate values X and Y of the two front and rear track points into the formula (1) above to find the distance s between the two points. And the time t consumed by moving between two points=current time (t ₁ )−last time (t ₀ ).

Substituting s and t into the above formula (2) can measure the speed at which the pen moves between two points. v=s/t

Then calculate the acceleration, v ₁ represents the speed at the current moment, v ₀ represents the speed at the previous moment, t ₁ represents the current time value, t ₀ represents the previous time value, according to formula (3), the current speed value, system time Value and the time and speed value of the last point in the recalled record sequence are substituted into formula (3), that is, the acceleration value of the pen between two points is calculated.

Acceleration = (v ₁ -v ₀ )/(t ₁ -t ₀ )

The value of Pressure is the same as the previous one, divided into 1024 levels. And store these index values in the memory memory sequence according to the law.

As the writing continues, the trajectory feature data generated by the movement is stored in the memory sequence one after another.

When the user enters the operation command of "writing completion", the integrity of the data in the memory sequence is detected, and then the speed, acceleration, average force value and fluctuation value of the entire trajectory are calculated according to the data sequence.

When the pressure value of a track point is lower than 100, and the values of the two points before and after it are both higher than 100, it is regarded as a fluctuation, and 1 is added to the fluctuation frequency counter.

According to the sum of the distances in the data sequence and the time from the beginning to the end, the average speed of the entire trajectory is calculated by substituting into formula (2).

According to the time and speed values at the beginning and end of the data sequence, the average acceleration of the entire trajectory from the beginning to the end point is calculated by substituting it into formula (3).

According to the pressure value in the data sequence, the average writing strength value of the entire trajectory is calculated.

After the calculation of the above data is completed, save the data to the database of the application program, so as to view and analyze in the next step.

The following is a specific calculation example:

1. The subject writes the first position point on the input board, the coordinates are (1526, 1728), the pressure is 542, the system clock is 8.256 (seconds), and the data are as follows:

2. Then write and move to the second position, the coordinates are (1673, 1790), the pressure is 600, and the system clock is 8.315 (seconds).

Calculate the distance from the first point to the second point:

$\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; =</span>\sqrt{{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1673</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1526</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1790</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1728</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 159.540</span>}$

Convert S from internal computer units to millimeters:

S=S/56.7

＝2.814(mm)

T=8.315-8.256=0.059 (seconds)

Speed from point 1 to point 2:

V=S/T

= 2.814/0.059 = 47.695 (mm/sec)

The acceleration from the first point to the second point, using (mm/s2) as the acceleration unit:

Acceleration=(v1-v0)/(t1-t0)

=(47.695-0)(8.315-8.256)

＝808.390(mm/square second)

Then the data stored in the sequence is:

Calculate according to this method, when the program receives the "writing completion" operation command, then calculate the velocity, acceleration, average force value and fluctuation value of the entire trajectory.

Claims (6)

1, a kind of handwriting analysis method, utilize computer program to realize handwriting analysis, it comprises the steps: (1) The subject uses the special pen for the input board to write on the surface of the input board connected to the computer, so as to obtain the input writing track; (2) Detect the validity of the writing track, filter out the invalid writing track, and keep the valid writing track; (3) obtain various feature data of each track point of effective writing track and record in the program memory, described feature data is the coordinate position, time point, distance, velocity value, acceleration value and pressure value of track point; (4) Calculate the average value of some characteristic data, and automatically select the characteristic data of multiple distribution track points according to the law, set up a data sequence reflecting the changing trend, and calculate the average speed, average acceleration, and average speed of the entire writing track according to the data sequence. Pressure value, fluctuation value and breakpoint number; (5) Display the specific values of various characteristic data, and give the change curve diagram of various characteristic data; (6) Analyze the writing track and process according to the specific values and change curves of the characteristic data. 2. handwriting analysis method as claimed in claim 1, is characterized in that, the data sequence that obtains is saved in storage database, can call out the record between different testers before and after the same tester as required, put in together for visual comparison. 3. The handwriting analysis method according to claim 1, wherein when the pressure value of a certain track point is lower than 100, and the values of the two points before and after it are both higher than 100, it is considered that there is a fluctuation , the fluctuation value is increased by 1. 4. The handwriting analysis method according to claim 1, wherein the average speed of the entire track is calculated according to the sum of the distances in the data sequence and the time elapsed from the beginning to the end. 5. The handwriting analysis method as claimed in claim 1, characterized in that, according to the time and speed values at the beginning and end of the data sequence, the average acceleration of the entire trajectory from the beginning to the end point is calculated. 6. The handwriting analysis method according to claim 1, characterized in that, according to the pressure values in the data sequence, the average writing pressure value of the entire trajectory is calculated.

Images